VALAR WATCHTake action

Origins

Founding, entity, technology and its lineage, and the milestones the company announced, set beside the record.

recordconfirmed Between March and July 2026 Valar's NRC-suit post, 'CAT scan' claim intact, was tagged asking search engines not to list it
2026 · orig-061 · Valar Atomics

Valar's April 7, 2025 post 'Valar Atomics is Suing the NRC' contains the claim that holding Ward One's spent fuel for five minutes gives the same exposure as a CAT scan (orig-017). Read in late September 2026, the page carries a robots meta tag, 'noindex, nofollow', which asks search engines not to list it, and the site's page data marks it 'disableIndex: true'; no other Valar library post carries the tag. Internet Archive copies of the page from April 8, 2025 through March 23, 2026 have no such tag; copies from July 22 and August 25, 2026 have it. In every copy the claim's text is unchanged, and no correction or note has been added. The post is still listed in Valar's own library, so visitors to the site can still find it. The tag appeared some time after March 23, 2026. The record does not show who added it or why.

In plain termsValar's post with the 'CAT scan' claim was still on its website, word for word, in late September 2026. Sometime between late March and late July 2026 the page was tagged asking search engines not to list it, while it stayed on the company's own site. No correction or withdrawal of the claim was found. No statement of why the tag was added was found in public records as of late September 2026.
verified 2026-10-02: Second check, late September 2026: live page and Internet Archive captures of 8 Apr 2025, 24 Aug 2025, 22 Dec 2025, 26 Feb 2026, 23 Mar 2026 (no robots tag) and 22 Jul 2026, 25 Aug 2026 (tag 'noindex, nofollow') re-read; the CAT-scan sentence is identical in all; library data marks only this post 'disableIndex: true'. Match. Corrected in late September 2026: the headline now says the tag asks search engines not to list the post, and the absence of a correction is stated as 'no correction found'. Reworded 30 Sep 2026: the detail and the plain-language text. Of the captures named, only those of 23 March and 22 July 2026 are kept, with the live page; no copy of the others is kept.
analysisconfirmed Carbon site in brief: state lease for a nuclear campus approved; no build permits found; a Valar mining claim 2 mi south
2026-09 · csite-001 · Valar Atomics, Utah Trust Lands Administration, Carbon County

As of late September 2026: the Trust Lands board voted unanimously on 17 September 2026 to approve an 'Other Business Arrangement' with Valar Atomics on Section 16, T14S R11E, about five miles north of Wellington: 480 acres now, the other 160 if a National Guard lease is moved, 50 years plus two 25-year extensions, with an option to buy, for a $400,000 upfront bonus and, in the Trust Lands slides' words, 'Approximately $35,000,000 in lease rents over 50 years' (csite-002, csite-004, mine-011). The stated purpose is an integrated nuclear energy, advanced manufacturing, critical mineral refining and data center campus: high-temperature gas-cooled reactors, TRISO fuel manufacturing, hydrogen and synthetic fuels, desalination and steel (csite-007). Separately, a Notice of Location names Valar Atomics Inc as locator of the 'RW 453' lode mining claim in Section 28, two miles south, located 12 September 2026; it is not yet in BLM's public claim layer, which the 90-day recording window allows (csite-019, mine-453, dir-005). No county permit application, state DEQ permit or DOE/NEPA document for the Carbon site was found (csite-012, csite-020, csite-023). Whether any NRC filing exists could not be checked, because NRC's website could not be read (csite-022), and water-right filings were not searched (mine-017). The only official actions found are the board's vote and the commissioners' 29 June support letter, which was not found on any Carbon County Commission agenda posted for its 6 May to 16 September 2026 meetings (mine-013, csite-026). No departure from the notice rules read was found (csite-005, csite-016, csite-018), but the first public document found naming Valar for the site was posted on 10 August 2026, and the April state review notice did not say nuclear (csite-032). Open questions: waste-storage framing (csite-008), the NRC route vs the CEO's 2025 stance (csite-022), a one-test NEPA review vs a larger build-out (csite-021), officials' praise of Valar's openness beside county agendas that did not name the site (disc-031) (csite-028), and a lode claim with no mineral stated (csite-019, disc-003).

In plain termsIn short: the state has agreed to rent Valar about three-quarters of a square mile north of Wellington (the whole square mile later if a National Guard lease is moved, with an option to buy) for a future nuclear and industrial campus. Building must wait until Valar has all required permits, and no application for those permits appears in the records we could check. The deal followed the notice rules we could check, but under those rules the first public notices named neither the company nor a nuclear use, and the first public record found naming Valar for the site came about five weeks before the vote. Separately, a Notice of Location shows Valar has staked a mining claim two miles south. The next steps (a county permit, water rights, state and federal nuclear approvals) are where the public may still weigh in; how much depends on the county's rules and on which federal route is used (csite-015, csite-020).
verified 2026-10-17: Re-checked in late September 2026 against the September memo (pp. 18-27), the 17 Sept transcript (motion; 'Is unanimous') and each cited fact. Corrected: the earlier 'Valar holds no mining claims there' is superseded by the Notice of Location (mine-453); 'critical mineral refining' added from the memo summary; NRC filing status now stated as unchecked, not as absent. Corrected in V1.1 (October 2026): all required permits must come before any construction, as the lease memo says; see Corrections.
recordconfirmed None of Valar's public pages or published technical documents mention the Carbon County site (as of late September 2026)
2026-09 · csite-027 · Valar Atomics

Read in late September 2026: Valar's home, Ward 250, Technology, Mission, Careers and Castle Country pages; its Library page (five company posts plus media links); and the Ward 250 Nuclear Safety Design Agreement and QA Program Description in its 'Public Documents' folder. None mentions Wellington, the Trust Lands lease, Section 16 or a Carbon County campus; on the Castle Country page 'Carbon' appears in its 'Carbon & Emery County' headings and history and as a partner logo (extends mine-020). The company does state its general plan there: 'Valar Atomics is building America's first nuclear gigasites', which it describes as energy campuses powered by fleets of reactors, without naming a location. The same page says 'Our doors remain open through tours, town halls and straight answers.' As of the pages read in late September 2026, the only Valar statements about the Carbon site found were its staff's remarks to the Trust Lands board (mine-018) and to the Carbon County Commission, whose 19 August minutes do not name the site (mine-013). That evening, at Valar's town hall in Price, KOAL News reported its chief executive saying Valar would expand onto BLM land 'if and when we outgrow the SITLA land', and ETV News reported Valar officials saying 'the 640-acre parcel is only the first step in its plans'; on 6 October he told the Carbon County Planning Commission that Valar would not build if the community did not want it (pcm-001).

In plain termsThe company tells residents on its local page that it offers 'straight answers', and it says there that it is building nuclear 'gigasites', but none of its own public materials read in late September 2026 says the state has approved leasing it land north of Wellington for a nuclear campus. In the records read, the site was first named in public in state board records, and the first local news report found came after the vote.
verified 2026-10-17: Holds, with the company's own 'gigasites' line added for balance. Re-read in late September 2026: home, ward-250, technology, mission, careers, castle-country and docs pages; QA PDF (24 pp.) and NSDA PDF (56 pp.) text-searched for Carbon, Wellington, trust land, Section 16, gigasite: none.
analysisconfirmed Funding in brief: over $1B of private money, $0 federal cash, big state offers, and key numbers not found in any filing
2026-09 · fund-001 · Valar Atomics, U.S. Department of Energy, State of Utah, Sequoia Capital, Erebor Bank · Series B equity (company-announced): $1,000,000,000; federal cash obligated to Valar (USAspending): $0; maximum Utah tax credit offered: $106,737,499

Holds: DOE's rule that pilot companies get no federal money matches the spending record ($0, fund-011); the $1 billion Series B is confirmed by Valar and its counsel, and Sequoia's portfolio page lists Valar (fund-007); the July 4, 2026 criticality goal given to investors and the public was met (fund-020). Where claims and record part: the seed round is $19M in the press and $18M on a participating investor's own blog (fund-003); every valuation ($2B, $6B) comes from Bloomberg as relayed by others, never from Valar (fund-006, fund-007); round totals overlap, so the equity raised is a range, about $1.15-1.5B (fund-008, fund-023); the February 2025 seed coverage cited a Philippine reactor contract, and no reactor there was on record as of the latest report read (April 2026; fund-003, fund-020); an investor quoted in an investor-published profile projected $1B a year of revenue per 100 reactors, and no revenue or customer contract was found in public records as of late September 2026 (fund-019). On incentives: in April 2026 an Emery commissioner said Valar had 'not even asked' the county for a tax incentive; in July the state approved up to $106.7M, which still needs a local incentive that had not been found on Emery County agendas through September 2026 (gov-033, fund-017). Not found in public records as of late September 2026: any Form D by Valar for any round (fund-010); the lender and terms of the reported $110M of debt, and the terms of the $200M facility (fund-006, fund-009); how that lending was handled given that the OCC's 2025 approval letter for Erebor Bank, which leads the facility, lists Valar backer Palmer Luckey as a director and principal shareholder, a tie neither announcement mentions, though nothing read shows improper terms (fund-009); the value of federal help in kind (DOE review at no charge, HALEU fuel shipped from a federal site with ownership not found in public records, a Los Alamos test, a possible Price-Anderson indemnity; fund-011 to fund-014); the terms of Valar's use of the public lab (fund-015); and any money set aside to decommission Ward 250, whose safety agreement marks the 'Financial Qualifications' chapter 'Not required for application' (fund-018). Carbon County: no incentive found as of late September 2026 (csite-029); the approved Trust Lands lease terms call for a $400,000 bonus and yearly rent (fund-016).

In plain termsValar has raised well over a billion dollars from private investors, including major Silicon Valley funds and people from defense-technology companies. It has received no cash from the federal government, which matches DOE's program rules, but the government helps in ways that have no published price tag, and Utah has offered up to $107 million in tax credits paid out of future new taxes. Many key numbers come only from press reports: the standard SEC notice that would show how much Valar raised and who runs it was not found on the SEC's filing system, and its loan terms were not found in public records (as of late September 2026). A reactor announced for the Philippines had not been built as of the latest report, and no revenue was found on record as of late September 2026, though an investor projected large future sales; Valar's July 4 reactor deadline was met.
verified 2026-10-17: Re-verified in late September 2026: summary rechecked against each linked fact; corrected 'lead investor' (Initialized was a participant, Riot led), 'federal HALEU' (ownership not found in public records), the $200M facility's lenders (named; only terms unknown) and the April 'not even asked' remark (it concerned the county).
analysisconfirmed Government connections in brief: what holds, where officials' words and the record part, and what is still not found in public records
2026-09 · gov-035 · Valar Atomics, U.S. Department of Energy, State of Utah, Utah Office of Energy Development, Emery County Commission, Carbon County Commission, Utah Inland Port Authority

Holds on the record: DOE did select (12 Aug 2025) and authorize Ward 250, the only one of DOE's four summer-2026 criticalities outside a national laboratory (gov-006, orig-010); DOE's agreement with Valar carries $0 of federal money, like all nine pilot agreements on file at USAspending (gov-003); Valar's federal lobbying is modest, disclosed and aimed at Congress (gov-004); Emery County's land sale followed state statute (gov-022); Valar's statement that DOE reviews and approves its design holds for the test reactor (gov-039). Where statements and records part: KUTV reported the governor speaking of NRC oversight while the state's signed MOU named DOE rules and Utah was suing to keep small reactors out of NRC licensing (gov-012, reported); the state lab's FAQ still tells the public 'the planned reactor' gets NRC licensing and NRC-tested operators, which the lab's only operating reactor does not have (gov-026); the state lab calls 18 June 'full-power' criticality when DOE's release and a DOE slide prepared for the NRC describe it as zero power (orig-012); Valar told the state's incentive board of plans for more commercial units and behind-the-meter sales, which the DOE route does not cover (gov-033, gov-039); Valar said in April 2026, at the Emery parcel hearing, it would 'be open with all plans'; the first public record found tying it to the Carbon site is the August 2026 agenda (gov-038); DOE counted reactors at Idaho National Laboratory toward a goal titled 'outside the National Laboratories', though its own application rules allowed lab sites (gov-006); a think tank's release called NOVA the pilot's first criticality, DOE says Antares was (gov-028); officials' statements on transparency and the MOU's confidentiality term and its bar on publicity without both parties' consent sit side by side in the record (gov-011, gov-020); the DOE-Utah campus pact cites Ward 250's 'successful permitting', but no state permit was found as of late September 2026 (gov-030); county and Inland Port records disagree on whether Valar's land is in a tax-capture zone (gov-023); Valar's groundbreaking date differs from the state's (gov-015). Not documented in the public records read: where Valar's HALEU came from, given it is in none of DOE's announced allocation rounds (gov-008). Not found in public records as of late September 2026: the OTA text, the DOE safety approvals themselves (a DOE slide prepared for the NRC Commission's 21 July 2026 meeting, found on 17 October 2026, lists the PDSA and DSA approval dates, February and 23 April 2026, but is not an approval document) and the readiness report Valar described to the state board in May 2026 (DOE confirmed in July 2026 that a federal readiness review covering Ward 250 ended in a final report to the startup authority; safe-007, gov-037, dir-031), and the state lab use agreement and board-approved terms (gov-013). The pilot program's application rules are posted on the FedConnect portal, which was not read (gov-032). Process: no Federal Register notice or comment period specific to Valar was found as of late September 2026 (gov-029), no local-government consent required under the pilot's rules as summarized (gov-032), and local approvals given quickly, sometimes with no review recorded (gov-021, gov-022). None of this record shows intent; it shows gaps that records requests can close.

In plain termsOn the government side, the basic facts check out: the Energy Department approved the reactor, it paid no cash for it, and the local land sale followed the law. In several places officials' descriptions of oversight and openness differ from their own papers: the NRC was cited as a safety check it never became for this reactor, a state web page still describes NRC licensing, and the state's May 2025 MOU with Valar carried a confidentiality clause when its lab director told residents in June 2025 'We are not trying to keep anything from anyone.' Valar's April 2026 'open with all plans' was said at the Emery parcel hearing; the first public record found tying Valar to the Carbon site is the 10 Aug 2026 agenda. Money and fuel questions remain: who supplied Valar's special uranium, and what the government's free safety reviews cost. No hearing or comment period on the reactor's safety case (its safety analysis or DOE's approval of it) was found at any level as of late September 2026. Emery County held a general public hearing on nuclear industries in June 2025, where residents raised safety concerns (gov-020), and state meetings that took up the lab and Valar had general public-comment periods on their agendas (gov-040, gov-041, gov-042). The open questions list says exactly which records would settle each point.
verified 2026-10-17: Summary rewritten in late September 2026 to match the verified facts: gov-012 now 'reported'; the FAQ, DOE commercial-licensing and INL-siting nuances added; new facts gov-037 to gov-039 included. Corrected in late September 2026: openness comparisons restated with dates and context; headline absence now worded as 'not found in public records'; restored the MOU's 'without both parties' consent' qualifier (gov-011, gov-020). Corrected 2026-10-01: wording made more exact against the cited records.
analysisconfirmed Local record in brief: a Carbon plan was public in 2025, the site was not; much talk in Emery, few chances to object
2026-09 · loc-033 · Valar Atomics, Carbon County, Emery County, Orangeville City, Utah Trust Lands Administration

As of late September 2026, testing against local records the view that the Carbon site should have been handled in public and that residents did not know. What holds for the company: in Emery County Valar appeared before towns, districts and residents at least 15 times and held open houses (loc-027); Emery's land sale was noticed by name and heard (loc-007); its fuel plant was disclosed in public (loc-016); and in 2025 the local paper printed Valar's 'gigasites' aim and reported a plan for a Carbon County manufacturing hub, without a site (loc-034). What the record shows on the Carbon site: the state's newspaper notice named no applicant or use (loc-032); no report or minutes of Valar's June-September appearances before Carbon bodies, including a Price council work session, name the Wellington site (loc-012); the local paper first reported it the day after the vote (loc-011); a week before the vote the CEO spoke publicly of hundreds of reactors in Carbon and Emery without, as reported, naming a site (loc-001); a commissioner said project NDAs are usually companies' demand (loc-002), and the county's own minutes record it kept a 'strict' NDA on a different factory prospect until August 2026 (loc-036); no NDA with Valar was found in the records read. Gaps between words and records: the state lab director, as ETV reported, said firms must show 'no dose' at the fence, while Valar's own analysis gives a small dose (loc-035); Carbon commissioners, as reported, called the test reactor 'commercial' and the data center waterless (loc-005); a local report of a chamber tour said meltdown risk is gone (loc-006); Valar told Castle Dale its reactor needs about 10 acre-feet of water for industrial use while DOE's review lists only domestic use (loc-025); officials and the state port authority describe the Emery land as sold while Valar said in August it awaited clean title (loc-014); Valar said it 'relocated' from California while most of its openings are there (loc-029); in April Valar said many of its 120 Emery workers did not live in Utah (loc-004). Process gaps: district minutes record a staff report that Valar tied into the sewer 'without prior authorization' (loc-015); the state declined to review the reactor building's fire plans and no record read shows who did (loc-018); Emery had no nuclear zoning rules and has now paused new major permit applications (loc-024); Orangeville is drafting a zone for Valar's land, its drafts to go through the city's and Valar's attorneys, with no hearing on it found as of 17 October 2026 (loc-013); no water-right application by Valar was found among the State Engineer's notices in the local paper (loc-031). None of this shows intent; it shows where the public was and was not given the facts in time to act, and which records requests would close each gap (open_questions).

In plain termsIn Emery County, Valar talked to people a lot, and some of what it and local officials said, as reported, differs from the documents. In Carbon County the general idea was public: the local paper reported in 2025 that Valar planned a factory hub there. But the specific plan north of Wellington was handled by the state in Salt Lake City; the local ad did not say what it was for, and Valar's local talks before the vote did not mention it, and the local paper first reported the site the day after it was approved; before then, the records we found name Valar for the site only in the state board's online notices and papers, from 10 August, and at its meetings in Salt Lake City. The formal moments when residents can object, a county permit hearing, a water-right protest, a zoning hearing, have mostly not happened yet. Those are the next chances to take part.
verified 2026-10-18: Rewritten after the earlier check: adds the 2025 local reports of a Carbon plan (loc-034), the lab director's dose statement (loc-035), the county NDA record (loc-036) and the Ferron staffing figure; corrects the fire item and the moratorium status. Each point rests on the fact it cites. Revised before publication (late September 2026): sewer, dose and comparison lines restated.
recordconfirmed Valar's public document trail: nine web posts, four unlisted, two technical PDFs; the safety analyses are not among them
2026-09 · orig-071 · Valar Atomics

Read in late September 2026, the page data behind Valar's Ward 250 page lists nine library posts. Five are listed: 'Valar Atomics is Suing the NRC' (April 7, 2025), 'Roadmap to the American Nuclear Golden Age' (May 23, 2025), the DOE selection post (August 12, 2025), the Project NOVA release (November 18, 2025) and the Series B post (August 4, 2026). Four are not listed but load at their addresses: the Ward250 NSDA post and the QA Program Description post (both dated October 13, 2025), the 'Ward 250 By the Numbers' draft (June 28, 2026, with placeholders) and the 'Ward Zero Whitepaper' (orig-058). The only technical documents found on the site are two PDFs in the site's 'Public Documents' asset folder, the NSDA (document 100403, rev 02, marked Draft) and the QA Program Description (document 100474, rev 02), both linked only from the unlisted posts. Three placeholder posts seen in April 2025, with 'Lorem ipsum' bodies, are gone. Nothing on the site publishes the PDSA, the final DSA, the Technical Safety Requirements, DOE's safety evaluation, the readiness review, the startup approval, the OTA or any Ward Zero test data (safe-007, orig-055, orig-059). The press list's newest entry is dated February 28, 2026 (orig-048), and the Ward 250 page's progress list still stops at the September 2025 groundbreaking.

In plain termsValar's own library of posts and documents fits on one short list: nine web posts, four of them not listed on the library page, and two technical PDFs, one marked 'Draft'. The documents that would show whether the reactor is safe as built, and what DOE required, are not among them. None of the other Valar pages read in late September 2026 links to the four unlisted posts.
verified 2026-10-15: Second check, late September 2026: library and Ward 250 page data re-read: nine posts, four unlisted (NSDA, QA, By the Numbers, Ward Zero Whitepaper), only the NRC post marked noindex; the two PDFs are linked only from the unlisted posts' bodies; NSDA marked 'Draft'; April 2025 archive shows three 'Lorem ipsum' posts, gone by October 2025. Match. Revised before publication (late September 2026): listing wording now 'unlisted'/'not listed'; source title parenthetical now 'unlisted'.
analysisconfirmed Origins claims vs record, in brief: what holds, what does not, what cannot be checked (as of late September 2026)
2026-09 · orig-074 · Valar Atomics

Does not hold, or conflicts with the record: 'first company ever to take a nuclear reactor critical outside of a national lab' (General Atomics did so in 1958; ppl-030, orig-066); electricity 'just a week later', actually thirteen days by Valar's own dates (orig-070); the spent-fuel 'CAT scan' claim, which engineers calculated to be a fatal dose as press reports describe (their inputs, such as distance and cooling time, are not in the record read; the worked check on the Proofs page, pf-007, gives about 75 to 334 mSv whole-body for all the spent fuel held 50 cm from the body a day after shutdown, and less than a CT scan for one pebble, on partly assumed inputs; the two rest on different assumptions and are not directly comparable), still posted and since tagged to ask search engines not to list it (safe-026, orig-061); fuel listed as 4.95% in the safety agreement but HALEU in DOE's shipping review (orig-040, disc-001). Official statements that did not match the record: Utah's lab page calling June 18 'full-power' criticality (orig-012); the governor's 2025 description of NRC oversight (orig-065). Not borne out as of 2026: 'we are building and testing this reactor in the Philippines' (April 2025), with no reactor built there a year later (orig-018, orig-064). Holds only narrowly: 'first startup in history to split the atom' (a Los Alamos-run test; The New Republic reported in May 2026 that Taylor walked the claim back after it was pointed out that other venture-backed companies had done it years earlier, though Valar's Castle Country page still carried it on September 29, 2026; orig-047); 'record time' (true from the Utah groundbreaking, not for the reactor hardware; orig-060); 'no burden on our water supply' (true for reactor cooling, not the wider campus; orig-070). Holds: DOE selection and criticality before July 4, 2026 (orig-006, orig-010); 'seven months' to criticality, counted from the Nevada core test (orig-015); 'first-ever C-17 airlift of a nuclear reactor' as worded (orig-066); the unfueled airlift (orig-030); the funding announcements, as reported (orig-057); Delaware incorporation (orig-051). Unsupported in public: Ward Zero test results the safety case relies on, with the only 'whitepaper' page holding the text of a 2019 essay by another author, not Ward Zero data (orig-058, orig-059); hydrogen and synthetic-fuel production (orig-063). Cannot be checked yet: the July 4, 2023 founding date (orig-053); the power Ward 250 reached (orig-013); 'world's first binding Coordinated Research Project' with PNRI (orig-064); that Ward One is named in the NRC lawsuit (safe-028); who made the fuel (orig-032). Changed over time: from a 12-month test in a temporary facility to a long-term fuel-and-reactor base and a Carbon County campus (orig-045, mine-019); from fuel synthesis to AI power (orig-063). Public dealing on Carbon County: as of late September 2026 no Valar publication was found that names the Wellington site or the claim it staked; its staff spoke at the Trust Lands board's August 20 and September 17 public meetings that took up the lease, and on the evening of September 29 Valar held a town hall in Price on the site (orig-075). Each gap is stated as a gap; none of this shows intent.

In plain termsPut side by side with the documents, Valar's origin story is a mix. Its biggest milestones did happen: DOE picked it, the reactor went critical before July 4, and the airlift took place. But several of its claims are wrong or stretched, a spent-fuel safety claim that outside engineers, as reported, calculated to be badly wrong, and that a worked check using the post's own figures, with the missing inputs assumed, finds true only for a small piece of fuel or after long cooling (pf-007), is still online, a Philippine reactor it said it was building had not been built as of April 2026, by Philippine reporting, and key evidence (Ward Zero test data, the safety analyses, the DOE agreement) was not found in public records as of late September 2026. The pattern is claims that run ahead of the public record; none of it shows intent.
verified 2026-10-15: Second check, late September 2026: summary rebuilt from the facts as verified today: 'seven months' moved from 'does not hold' to 'holds' (it matches NOVA to Ward 250 criticality); the Philippine plan moved to 'not borne out as of 2026' (a stated plan, not yet built); state and governor statements listed separately from Valar's own claims. Revised before publication (late September 2026): whitepaper and search-tag wording restated. Revised before publication (2026-10-01): wording made more exact against the cited records.
analysisconfirmed Plans and operations in brief: one small test reactor ran; the far bigger plans exist as words, a lease and job ads
2026-09 · plan-001 · Valar Atomics, U.S. Department of Energy

What holds (as of late September 2026): Valar built a 100 kWt test reactor, heat-tested it in California and flew it unfueled to Utah; DOE approved HALEU TRISO fuel for shipment to it from a federal site in Nevada (no record found states what fuel is loaded, disc-004); and DOE announced its zero-power criticality on 18 June 2026, ahead of the July 4 goal and of Valar's own 30 June target (plan-002, plan-004, plan-008, plan-014). A fuel lab stands beside it and is hiring for a pilot TRISO line (plan-015). What rests on a DOE listing without data: full power, which a DOE slide prepared for the NRC Commission's 21 July 2026 meeting lists on 2 July 2026 (plan-006; the only data is an undated Valar chart with no power units, plan-007). What rests on the company's word: the electricity demonstration's output (ETV reported the reactor at 37% power, plan-010), and the fuel plant's future size (plan-015). Where dates clash: full power on 'June 21' in the state's lease memo, which does not say who supplied the date, versus a 10 kWt output Valar announced 22 June and a 2 July full-power date on a DOE slide (plan-009, plan-006). Where Valar's own statements clash: a 12-month demonstration in the approved papers versus a permanent base and 'largest' fuel maker in local talks (plan-003, plan-015); a July 4 date still in the future tense on the homepage (plan-019). Where a Valar document and a DOE document differ: the fuel is listed at 4.95% enrichment in Valar's safety agreement but as HALEU (under 20%) in DOE's shipping review (plan-014, disc-001). What is plan only: fleets and 'thousands of reactors per year' (plan-020), a Carbon County campus with no reactor count, size or date in the state's lease papers (plan-016), and 'Project Beehive' near Price, which NPR reported on 30 September, from a Valar proposal to federal regulators, as some 456 small reactors with the first in 2028 (plan-017), a 30 MW NVIDIA data center (plan-010), a turbine program (plan-018), 'one of Valar's first commercial projects' in Utah (plan-011). What has not happened: the Philippine reactor, the gas-fired test heater, fuel from Valar's own line, the Green River gigasite (plan-021). No power buyer, utility contract or military reactor contract is on the record (plan-012, plan-013). Every site on record is in plan-017; every physical movement of reactor and fuel in plan-025.

In plain termsValar did what it said it would do first: it built a small test reactor and got it running in Utah before July 4, 2026. Almost everything beyond that is still a plan: fleets of reactors, a Carbon County campus, a data center, a fuel factory at full scale. Some of the company's own dates and descriptions disagree with each other, and the government's public announcements cover only the first step, though a DOE slide shown to federal regulators in July also lists full power on 2 July. None of this shows danger or bad faith by itself; it shows how much of the story has no public record behind it yet.
verified 2026-10-17: Checked in late September 2026: summary re-checked against the corrected facts it cites; fuel wording changed from 'shipped' to 'approved for shipment' (no shipment record was found); fuel-enrichment difference added. Revised before publication (late September 2026): 'fuelled it with HALEU' now approved for shipment; no public record states what fuel is loaded.
recordconfirmed California is the factory: Valar says Ward 250 was heat-tested in Hawthorne; 81 of 101 job posts were in Torrance on 17 Oct 2026
2026-09 · plan-008 · Valar Atomics, U.S. Department of Energy

Valar's unlisted, unfinished post dated 28 June 2026 says Ward 250 underwent 'extremely rigorous non-nuclear testing' at its headquarters in Hawthorne, California, before it reached Utah, with temperatures above 750 C and pressures over 6 MPa, thermal cycling and transients; the whole reactor system is six high-cube shipping containers. DOE's Fuel Line Pilot Program page lists the company as 'Valar Atomics Inc. (Hawthorne, California)'. On 29 September 2026 its public job board listed 71 of 92 openings in Torrance, California (on 16 and 17 October 2026, 81 of 101), including a graphite production lead for machining 'reactor components', CNC programmers, machinists, a welding lead for welds to 'nuclear-grade specifications', and eight turbomachinery roles; two posts name Hawthorne as a work base (a Torrance-listed commercial project manager to 'operate in both Hawthorne and Utah', and an Orangeville-listed people-operations director to be based in or relocate to Hawthorne), and a Utah post mentions travel to the 'Los Angeles headquarters'. Read on 16 and 17 October 2026, the board no longer listed the people-operations director post; the commercial project manager post and that Utah post were still listed. The Carbon County claim notice carries a Torrance address (orig-036). In short: reactors and their parts are designed, machined and tested in the Los Angeles area; Utah is where they are fuelled and run. Non-nuclear heat tests are not limited by the nuclear safety limits in the NSDA (a 650 C limit versus 750 C goals, disc-002), so the 750 C figure is not a conflict with them.

In plain termsValar says the reactor was tested without uranium at its headquarters in the Los Angeles area, then flown in pieces to Utah, where the uranium was added. Most of Valar's hiring is still in California: machinists, welders and turbine engineers. Utah is where the nuclear part happens.
verified 2026-10-17: Checked in late September 2026: feed re-counted (92 posts: 70 'Torrance, California' + 1 'Torrance, CA', 21 Orangeville; 8 Power Conversion turbomachinery roles). Fixed: only one Torrance-listed post names Hawthorne; the other is Orangeville-listed. DOE fuel-line page date set to its last modification (2026-02-12; first published 2025-07-14). Job boards change daily; counts as of late September 2026, re-counted 17 October 2026 (101 posts: 81 in Torrance, 20 in Orangeville).
recordconfirmed Every U.S. Valar site on record (late September 2026): one test reactor, one approved lease, three pending BLM files
2026-09 · plan-017 · Valar Atomics, Bureau of Land Management, Utah Trust Lands Administration, Emery County

Built or operating: Ward 250 and its fuel lab at the San Rafael lab, Emery County (plan-002); the company's design, machining and test shops in Hawthorne/Torrance, California (plan-008). Used once: the NOVA test core at the National Criticality Experiments Research Center, Nevada, operated by Los Alamos in November 2025 (orig-008). Being acquired: the 112-acre Emery County parcel beside the lab (gov-022, loc-014). Lease approved: on 17 September 2026 the state trust-lands board approved a lease with purchase option for 480 acres of the section north of Wellington, with the other 160 acres to follow if the National Guard's lease there is moved (mine-011, plan-016); no record read shows the lease signed. Applied for: three Bureau of Land Management cases (the two named ones in the Price Field Office), all 'Pending' in late September 2026: UTUT106784007 'Green River Gigasite Small Modular Reactor/Data Center' (10,270.61 acres, case date 13 February 2026, which Valar told The Washington Sun it no longer pursues, mine-008), UTUT106801248 'Project Beehive Small Modular Reactor/Data Center' (case date 13 April 2026, no acreage shown), and UTUT106833224, a distribution-line case with no name (entered 23 July 2026). NPR reported on 30 September 2026, citing a Valar proposal to federal regulators that NPR reviewed, that Project Beehive would cover over 9,000 acres of BLM land near Price with data centers, some 456 small reactors of 25 megawatts of electricity each, a fuel facility and nuclear-waste storage, with the first reactors in 2028; BLM's Utah office told NPR it had received an application and was reviewing it for completeness. The proposal itself was not found in public records. No BLM rights-of-way case names Valar outside Utah, and its other public layers show no Valar claims, notices, leases or permits (mine-005, mine-030). Valar says it looked at more than 20 states (mine-018, ppl-006, loc-001); no site in another state was found in any public record read.

In plain termsHere is the map of Valar's sites and land applications as far as public records go; a mining-claim notice naming Valar near Wellington is a separate record (mine-453). One small test reactor is running in Emery County, next to a fuel lab. The reactors are built in the Los Angeles area. Valar is buying land next to the Emery lab and has state approval to lease most of a square mile of state land in Carbon County. It has three pending cases with the federal land agency: the Green River site (over 10,000 acres; Valar says it dropped it), Project Beehive near Price (which NPR reports is planned for about 456 small reactors on over 9,000 acres) and a distribution-line right-of-way.
verified 2026-10-02: Checked in late September 2026: BLM's rights-of-way records re-checked for customer name Valar: three Utah cases, all Pending, the two named ones in the Price Field Office; Green River 10,270.61 acres, case date 13 Feb 2026; latest modification 15 Sept 2026. Fixed: lease wording overstated (board approval; initial 480 acres; signing not found); Launch Pad wording corrected. Revised before publication (late September 2026): earlier fix note reworded (lease wording). Checked 2 Oct 2026: the NPR paragraph matches NPR's 30 September 2026 article as published (456 reactors, over 9,000 acres, 25-megawatt units, first reactors in 2028, BLM reviewing the application for completeness). No copy of NPR's article or map is in the source archive. Corrected in V1.1 (October 2026): the detail says two of the three cases are in BLM's Price office; see Corrections.
company claimconfirmed Valar's homepage, Sept 29, 2026: July 4 still 'to go live', and 'we're building hundreds' of reactors on gigasites
2026-09 · plan-019 · Valar Atomics

Read in late September 2026, Valar's homepage says Valar is 'building hundreds of nuclear reactors on Valar Atomics gigasites' focused on hydrogen (by a sulfur-iodine cycle), data-center power, heavy industrial power and clean hydrocarbon fuels, and that it is partnering with DOE, the San Rafael lab and Utah 'to go live before America's 250th birthday on July 4th, 2026', a date now three months past, written as still ahead (Ward 250 reached criticality on 18 June, before that date; orig-010). Its technology page says the non-nuclear Ward Zero prototype is 'ready to go live'. The record: one test reactor has been built and operated (plan-002, plan-006); no gigasite has been approved anywhere (the Carbon lease holds land pending permits, plan-016; Valar says it dropped the Green River project, whose application BLM's public record still listed as pending on 29 September, mine-008); no hydrogen or synthetic fuel has been produced (orig-063). 'Building hundreds' describes an aim, not work under way. The newer Castle Country page uses past tense for Ward 250 and says 'Ward 250 is step one'.

In plain termsValar's front page still talks about a July 4 deadline as if it has not happened yet, and says the company is building hundreds of reactors at giant sites. In reality it has built one small test reactor. Nothing on the public record shows a giant site approved, or any hydrogen or synthetic fuel made. The page describes a goal in the present tense.
verified 2026-10-15: Checked in late September 2026: homepage and technology page re-read; all three quotes exact. Company pages are undated; 'published' is the read date.
analysisconfirmed Promised vs delivered, item by item (as of late September 2026): what held, what changed, what has not happened
2026-09 · plan-021 · Valar Atomics, U.S. Department of Energy, Emery County

HELD: 'First Reactor online by July 4, 2026' (Valar's July 2025 deck to Emery County): DOE announced criticality on 18 June 2026 (orig-010), ahead of Valar's own 30 June target (plan-004), and a DOE slide prepared for the NRC Commission's 21 July 2026 meeting lists full power of 100 kWt on 2 July 2026, without data (plan-006). HELD AS A PLAN, UNCHECKED: '30EFPD, in 12 months' (deck) matches Valar's June 2026 draft; no run log was found in public records as of 17 October 2026 (plan-006, plan-007). CHANGED: 'in a temporary facility' (deck): the approved papers still plan to take Ward 250 down after about a year of running (plan-003), but what was built is a 55-foot modular reactor complex on engineered foundations beside a fuel laboratory, with more than 4,000 cubic yards of concrete poured for the foundations of both (plan-002), on a site Valar now calls home and is expanding (plan-003, ppl-006); 'about 40 employees for about 12 months' of testing (8 July 2025 town hall, as ETV reported it) became a projection of 275 jobs and $1.3 billion for a state tax credit (loc-004, fund-017). LATE: DOE safety approvals came about two months behind Valar's own schedule, by the dates POWER reported and the same DOE slide lists (plan-004). NOT DELIVERED: the Philippine reactor announced in March 2025 (orig-064); the gas-fired supercritical-CO2 test heater Valar wanted done by June 2026, approved for design only (plan-018); fuel from Valar's own line: Ward 250's fuel was cleared by DOE to ship from a federal site in Nevada (plan-014). DROPPED: the 10,270-acre Green River gigasite, per Valar, though BLM listed it pending on 29 September; BLM's public layer, checked on 12 October 2026, shows the case closed as of 6 October 2026 (plan-017, mine-008). NOT YET DUE: 'Commercial Deployment by 2028' (deck), the 30 MW NVIDIA data center (plan-010), 'hundreds of reactors' in Carbon and Emery (ETV's paraphrase of Taylor, loc-001). On file so far are land filings, such as Valar's pending BLM application 'Project Beehive Small Modular Reactor/Data Center' in the Price Field Office (case date 13 April 2026, plan-017) and the Carbon County trust-land lease Valar sought, which the state board approved on 17 September (mine-011). CONFLICTING: the date of full power (plan-009) and, on paper, the fuel's enrichment (plan-014).

In plain termsChecked one by one: Valar kept its biggest public promise, a working test reactor before July 4, 2026. Several other things grew far beyond what locals were first told (a 'temporary' one-year test sits on a site the company now calls home, with a fuel lab beside it and talk of hundreds of reactors). Some promises have not happened (the Philippines reactor, a test heater, making its own fuel), and the filings found so far for the biggest future promises are land filings, such as a pending federal-land application for a reactor and data-center site near Price.
verified 2026-10-17: Checked in late September 2026: deck, NSDA Table 7, ETV/USREL update, Research Board minutes and BLM layer re-read. Fixed: the 'temporary facility' line now says the approved plan is still temporary; the delay figure corrected to about two months (plan-004).
analysisconfirmed 'Unmatched safety' and 'impossible to melt' tested: DOE says TRISO particles cannot melt; UCS says not meltdown-proof
2026-09 · safe-025 · Valar Atomics, Union of Concerned Scientists, Utah San Rafael Energy Lab

Valar's home page (read in late September 2026) says its HTGR architecture has 'an unmatched safety profile and proliferation resistance when paired with TRISO fuel'; the joint NOVA release calls Valar's reactors 'inherently safe'; the state lab's Valar page says the design is engineered to 'remain impossible to melt under extreme conditions' (orig-012). Valar's own NSDA is more measured: TRISO retains more than 99.99% of fission products in normal operation and keeps its integrity 'up to 1600°C during postulated accidents', with a failure fraction under 1 in 10,000 at 1600°C, a peak fuel limit of 1,250°C, and an accident analysis that assumes some particles fail. DOE's Office of Nuclear Energy says TRISO particles 'cannot melt in a reactor' and calls TRISO the most robust nuclear fuel (exp-001). The Union of Concerned Scientists' 2021 assessment of reactor types in general (it does not mention Valar) finds HTGRs have attractive safety features but that TRISO is 'far from meltdown-proof', releases fission products faster above about 1,600°C, depends on consistently manufacturing fuel to exacting specifications (not yet demonstrated, in its view), that HTGRs are vulnerable to air or water getting into the reactor, and that designs without leak-tight containment may cancel out the benefits; it also says using HALEU instead of lower-enriched uranium increases proliferation and terrorism risks, though far less than plutonium, and that TRISO fuel fabrication is harder to monitor than ordinary fuel fabrication. Ward 250 has no conventional containment by design (NSDA 'functional containment'), and its fuel shipped from Nevada is HALEU (safe-010).

In plain termsTRISO fuel is genuinely tough: each bead is its own tiny container, and tests show it holds in radioactivity to very high temperatures. But 'unmatched' and 'impossible to melt' go further than Valar's own safety paperwork, which sets temperature limits and assumes some beads can fail. An independent science group adds that the beads must be made consistently to exacting specifications, that air or water leaking into this kind of reactor is a known weak point, and that richer HALEU fuel is somewhat more attractive to people who might misuse it. 'Robust' is supported by the record, and DOE says the particles cannot melt in a reactor; but 'unmatched' and 'impossible to melt' go further than the company's own safety paperwork.
verified 2026-10-15: Re-read Valar home page, NOVA release ('inherently safe'), the state lab page, NSDA pp. 2, 8, 23 and the UCS executive summary in late September 2026. Holds. Fixed: the executive summary read names no author, so the attribution to Edwin Lyman was removed; noted it is a general 2021 assessment that does not mention Valar; plain no longer calls the claims 'marketing'. Cross-check in late September 2026: the headline read 'not meltdown-proof, per DOE limits and UCS', but DOE's own TRISO page says the particles 'cannot melt in a reactor' (disc-015, exp-001); 'not meltdown-proof' is UCS's assessment. Headline reworded to attribute each view; detail and plain unchanged.
analysisconfirmed Safety in brief: approved by DOE with no public hearing; public papers disagree with DOE's fuel record and each other
2026-09 · safe-038 · Valar Atomics, U.S. Department of Energy

Summary of this file as of late September 2026. Authority: Ward 250 runs on a DOE authorization under EO 14301 and DOE-STD-1271, with Valar as its own design authority and no public hearing step (safe-001, safe-002, safe-019); no public copy was found of the documents that define what it may do (DSA, TSRs, DOE's safety evaluation, startup approval) (safe-007); DOE says federal reviewers checked contractor safety submittals for the pilot reactors and took part in Ward 250's readiness review before startup (dir-031); and a DOE slide prepared for the NRC Commission's 21 July 2026 meeting (found 17 October 2026) lists the safety-analysis approval dates Valar gave (PDSA, February 2026; DSA, 23 April 2026), while, in the records found, the dates of the other DOE safety steps come only from Valar (safe-006). Where the record disagrees with itself or with Valar: DOE's shipping review describes HALEU fuel for Ward 250, and a DOE slide prepared for the NRC Commission in July 2026 lists its fuel as '19.75 wt% TRISO', while the only public safety agreement lists 4.95% fuel (safe-010, disc-001); power at 250 kWt is allowed for one hour per run in the NSDA and up to one day, possibly a total, in a later Valar draft (safe-008); the accident inventory assumes 3 MWd against a 30 MWd limit (safe-009); outlet temperature 650 C limit versus 750 C goal (disc-002); worst-case boundary dose stated as under 0.5 rem and under 0.1 rem (safe-013); a 400 m public boundary that cannot fit inside the 20.6-acre lab (safe-014); the NSDA's 400 m dispersion factor is 3.1 to 12.3 times lower than standard rural methods give for its own stated weather (safe-014); it cites 18 criteria of a QA rule that lists 10 (safe-018); its 18-month lifecycle is shorter than its own schedule plus 12 months of operation (safe-009); a state lab director's reported statement that any nuclear company must show no dose at the site boundary in an accident, versus the analysis (safe-015); '72 hours' versus 'indefinitely' of walk-away cooling (safe-020); and DOE's environmental review naming only a fault 150 km away when the state maps one about 15 km away, and saying no major emission source lies within 10 miles when a coal plant is about 5 to 6 miles away. Claims tested against the record: Valar's 'unmatched' safety goes further than its own safety agreement, which sets temperature limits and assumes some fuel particles fail; the state lab page's 'impossible to melt' is supported by DOE, which says TRISO particles cannot melt in a reactor, though that does not mean nothing can escape (safe-025, disc-015); the spent-fuel 'CAT scan' claim is still on Valar's site, with no supporting calculation or correction found there as of late September 2026 (safe-026). Public records not found: emergency plan, exercises and any DOE decision on on-site-only planning (safe-016), monitoring data (safe-024), incident reports (safe-033), spent-fuel destination and state permit (safe-023), the fuel line's authorization (safe-031), and the Utah-DOE agreements the NSDA cites (safe-034). Claims that hold are in safe-039. None of this shows the reactor is unsafe; it shows the public cannot check that it is safe.

In plain termsWard 250 was approved by the Department of Energy through a fast process with no public hearing. DOE says its teams reviewed safety papers for the pilot reactors and took part in this reactor's readiness review before startup, but no public copy was found of the papers that show exactly what it is allowed to do. The papers that are public disagree with each other in several places, including on the fuel: DOE's own shipping review, and a DOE slide shown to federal regulators in July 2026, describe richer fuel than the company's public safety agreement does. Some company claims hold up; some go further than its own documents; several promised safeguards (an emergency plan, monitoring reports, a destination for the used fuel) were not found in public records as of late September 2026. The fair conclusion is not that the reactor is dangerous, but that the public cannot yet check its safety case for itself.
verified 2026-10-18: Synthesis re-checked in late September 2026 against the verified facts it cites; updated for the fixes in safe-016, safe-019 and; each point is sourced in the fact it cites. Corrected in late September 2026: the safety agreement is now said to list, not analyse, 4.95% fuel; the 250 kWt limit now reads per run, the later figure possibly a total; a 14-day point was removed; the dispersion, 18-criteria and 18-month points were added; absence is stated as 'no correction was found'. Corrected 1 October 2026: wording made more exact against the cited records. Corrected October 2026: 'impossible to melt' is now attributed to the state lab page and set beside DOE's statement that TRISO particles cannot melt in a reactor; the advocacy-group assessment was removed; Valar's own wording 'CAT scan' is quoted, and 'disputed by engineers' was removed because the records this summary cites do not report it; the Valar-only dates are limited to Ward 250's DOE safety approvals in the records found; the state lab director's statement is given as reported speech; the plain text's absence line now carries its date; the state lab page, DOE's TRISO page, Valar's home page and Valar's lawsuit post were added as sources.
analysisconfirmed What holds up: DOE did authorize the reactor, the airlift was unfueled, DOE checked the fuel's shipping package, DOE rates TRISO fuel more resistant than ordinary fuel
2026-09 · safe-039 · Valar Atomics, U.S. Department of Energy

Claims tested and found supported by the record: DOE confirms Ward 250 is DOE-authorized and completed a zero-power criticality on 18 June 2026 (orig-010). The February 2026 airlift carried the reactor without fuel, and the fuel was to move separately from Nevada, consistent across World Nuclear News and DOE's packaging review (safe-012). DOE's packaging reviewers independently recalculated criticality safety for the fuel's shipping package and agreed with the applicant (safe-012). TRISO fuel's retention of fission products to about 1,600 C is stated in Valar's NSDA, and DOE says TRISO fuel is 'more resistant to neutron irradiation, corrosion, oxidation and high temperatures' than traditional reactor fuels. No active fault is mapped at the Ward 250 site, according to DOE's categorical exclusion and the Utah Geological Survey's fault map, and the same state map bears out Valar's no-fault claim for the Carbon parcel, as the Trust Lands memo relays it; the flood reasoning for Ward 250 is consistent across Valar's NSDA and DOE's categorical exclusion. The NOVA test was accurately described as operated by Los Alamos under NNSA oversight at zero power (orig-008). The 1 July electricity demonstration shows no sign of a commercial sale that would cross the EO 14301 line (safe-030). The company's passive-cooling design is consistent with its safety documents, though the June 26 test is known only from Valar (safe-020). Checked against Valar's own paper: its core-table arithmetic holds; its 100 m worker dispersion factor is at or above DOE's prescribed default; its 0.2 g earthquake value is close to the USGS design value for the site on soft rock (about 0.21 g); and its staffing of one senior operator and one operator per shift meets the NRC research-reactor benchmark.

In plain termsNot everything Valar says is contradicted by the record. The reactor really was approved by DOE and really did start up. It really was flown empty, and the fuel was approved to travel separately in a shipping package DOE checked for itself. DOE says the fuel type, TRISO, is 'more resistant to neutron irradiation, corrosion, oxidation and high temperatures' than traditional reactor fuels. Where the record supports the company, this file says so.
verified 2026-10-02: Re-checked in late September 2026 against DOE's 18 June 2026 release, World Nuclear News and DOE's packaging safety evaluation report. Holds. Revised before publication (late September 2026): added the NSDA values that check out (core table, 100 m factor, 0.2 g, staffing). Revised before publication (2026-09-30): reworded the detail; reworded the plain-language text; reworded the headline; removed the citation of Union of Concerned Scientists (advocacy publication); now cites U.S. Department of Energy, Office of Nuclear Energy instead; corrected the description of U.S. Department of Energy, Office of Nuclear Energy (quote).
analysisreported Valar and published investigations: three report no reply to requests for comment; a fourth interviewed Taylor twice and toured the site
2026-09 · ppl-029 · Valar Atomics, Isaiah Taylor, Kip Mock, The Utah Investigative Journalism Project, Mother Jones, ProPublica, The New Republic

How Valar and its people have responded to published investigative reporting, as the publications report it: the Utah Investigative Journalism Project (July 2025) says Valar 'did not respond to multiple requests for comment' and Kip Mock did not respond; Mother Jones (Feb 2026) says Taylor did not respond on several points, and that his promised write-up on the spent-fuel dose never came; ProPublica (Mar 2026) reports that 'Valar and its executives did not reply to requests for comment'; The New Republic (May 2026) interviewed Taylor in summer 2025 and again in early 2026; when its planned site tour with Taylor fell through (Valar told the reporter he had missed a flight), two Valar staff showed the reporter the site, and Valar's director of communication answered a question about an investor in one line; Jess Housekeeper told the reporter he had not read the ProPublica article and would not comment without doing so. No correction or reply to any of these pieces was found on Valar's website, whose spent-fuel claim remained online when read in late September 2026 (safe-026). Meanwhile Valar calls itself 'very open and transparent' (safe-040) and its staff speak of 'transparency to the public'. A company is free not to comment. Three outlets report no reply to requests for comment, and the fourth interviewed Taylor and was shown the site by Valar staff; no Valar correction was found on its site as of late September 2026.

In plain termsOf four published investigations, three report no reply from Valar to requests for comment; the fourth, The New Republic, interviewed Valar's chief executive twice and was shown the site by Valar staff. No correction of the disputed spent-fuel claim was found on Valar's website as of late September 2026. Valar describes itself as 'very open and transparent'.
verified 2026-10-16: Re-read the Utah Investigative Journalism Project, Mother Jones, The New Republic and ProPublica articles; Valar's NRC-lawsuit post with the spent-fuel sentence was still online in late September 2026. Valar's 'Ward 250 By the Numbers' page opened to confirm 'very open and transparent' (it still shows 'July XX' placeholders); added as a source. Corrected in late September 2026: headline, detail and plain restated to say what each outlet reports; Housekeeper's reply to The New Republic added; Valar's communications staff are described by role, not named. Corrected October 2026: ProPublica says Valar and its executives did not reply to requests for comment, so it is counted with the Utah Investigative Journalism Project and Mother Jones; The New Republic's two interviews with Taylor and its site tour with Valar staff are stated, and the one-line answer is credited to the person the article calls Valar's director of communication; the 17 September 2026 Trust Lands board transcript added as the source for 'transparency to the public'.
analysisreported 'CAT scan' spent-fuel claim still on Valar's site in late September 2026; no supporting calculation or correction found there
2026-09 · safe-026 · Valar Atomics, Isaiah Taylor

Valar's April 2025 lawsuit post, still online in late September 2026, says its analysis indicates that holding spent fuel from its 100 kWt 'Ward One' test reactor (planned life under a month) for five minutes gives radiation exposure equal to a CAT scan. Valar's Ward 250 safety paper, the NSDA, credits reactor shielding that 'cannot be inadvertently defeated', analyzes fuel-handling accidents, and plans defueling into storage; DOE's categorical exclusion has the spent core leave in dry storage casks. No Valar correction or published calculation was found on Valar's website as of late September 2026. For scale: a CT scan is about 1 to 10 millisieverts (FDA), so five minutes would need 12 to 120 millisieverts an hour; the post gives no amount of fuel, cooling time, distance or shielding.

In plain termsUsed reactor fuel is radioactive; how much depends on the reactor's power, how long it ran and how long the fuel has cooled. Valar said holding used fuel from its planned Ward One test reactor for five minutes would be like one hospital CT scan. No calculation supporting it was found on its website, where the claim remained in late September 2026. For Ward 250, its own safety paper treats radiation shielding as a safety function, and DOE's review has the used fuel leave in dry storage casks.
verified 2026-10-02: Re-read Valar's 7 April 2025 post in late September 2026: the five-minute CAT-scan sentence is still on the page, with no correction or supporting calculation. Corrected in late September 2026: added the CT-scan dose-rate scale; removed a quoted phrase that was not Valar's wording. Corrected 30 Sept 2026: reworded the headline, detail and plain-language text to rest on government and company documents; a news citation was removed; now also cites the U.S. Energy Information Administration and the U.S. Department of Energy, Office of Nuclear Energy.
recordconfirmed Where the site is, in the records' own words: Section 16, T14S R11E, about five miles north of Wellington
2026-09-17 · csite-002 · Utah Trust Lands Administration, Valar Atomics, Carbon County

The Trust Lands memorandum gives the location as 'T14S, R11E, Section 16: All*', Carbon County, 640 acres (+/-), for the Public Schools Trust, and places it 'approximately five miles north of Wellington'. The staff slides of 20 August 2026 label it 'Carbon County, North of Wellington' and a '640-acre TLA parcel' chosen by Valar 'following multi-state search'. The lease map is titled 'Special Use Lease Agreement No. 2095, Township 14 South, Range 11 East, SLB&M' (N2, SW4, 480 acres). The lease covers 480 acres at first (the N1/2 and SW1/4, mine-011); the SE1/4 is under a Utah National Guard lease (csite-003). Carbon County's 29 June letter calls it 'north of Wellington on SITLA lands' (mine-013). Staff told the board it is about six miles east of Price and six miles north of Wellington (20 August) and 'essentially six miles north of Wellington' (17 September), per the auto-generated transcripts; the memo says about five. ETV News, on 18 and 30 September 2026, gave the location as 'about six miles north of Wellington and six miles east of Price'. As of late September 2026, no street address or road name was found in the records read. The posted RW 453 claim notice is for Section 28, two sections (about two miles) south of Section 16 in the same township (csite-019).

In plain termsRural Utah land is located on a survey grid of one-mile squares called sections, grouped into townships and ranges. Section 16 of Township 14 South, Range 11 East names one exact square mile. The state owns it and holds it in trust to earn money for public schools. The records place it five to six miles north of Wellington, on open land away from town. The posted RW 453 mining-claim notice is for Section 28, two miles south of this square (csite-019).
verified 2026-10-17: Holds. Re-read in late September 2026: memo pp. 18-20 (location, key information), August slides p. 41, lease map p. 29, Carbon letter p. 28 (scanned), August 20 and September 17 transcripts. Added the staff's spoken 'six miles' next to the memo's 'five miles'.
recordreported Valar among four reactor developers carried into the second round of DOE's Nuclear Energy Launch Pad
2026-08-24 · orig-043 · Valar Atomics, National Reactor Innovation Center, U.S. Department of Energy, Idaho National Laboratory

World Nuclear News (Sept 4, 2026) reports that on August 24, 2026 DOE's National Reactor Innovation Center named 12 Launch Pad participants, including four reactor developers already supported under the initiative and the Reactor Pilot Program: Antares Nuclear, Deployable Energy, Oklo and Valar Atomics. It describes Ward 250 as a 100-kWt helium-cooled TRISO HTGR that reached zero-power criticality in June and was proceeding with power ascension. NRIC's director says the Launch Pad gives a prioritized path to DOE authorization plus expert and facility support; the article does not describe new funding for Valar. NRIC's own Valar page could not be read.

In plain termsIn August 2026 a DOE program for fast-tracking reactors kept Valar on its list. As far as the article says, it offers help and a priority path, not new money.
verified 2026-10-15: Second check, late September 2026: WNN re-read: Aug 24 selection of 12, four reactor developers incl. Valar, 100-kWt description, NRIC director on a prioritised path; no new funding described. Match.
recordreported Carbon County move: state trust-lands board agenda lists a 640-acre special use lease with Valar near Wellington
2026-08-20 · orig-044 · Valar Atomics, Utah Trust Lands Administration

The Utah Trust Lands Administration posted a notice that it had received an application, from an applicant it did not name, for Section 16, Township 14 South, Range 11 East, SLB&M, about 640 acres near Wellington in Carbon County, and invited competing bids to lease, buy or exchange it by August 7, 2026. The board's August 20, 2026 agenda lists a discussion item on a Special Use Lease Agreement (SULA 2095 OBA) for 640 acres in Carbon County with Valar Atomics. It was not the first record of Valar applying for land beyond the test lab: BLM's rights-of-way layer lists a 10,271-acre 'Green River Gigasite' case with Valar as applicant, created February 13, 2026 and shown as closed on October 6, 2026 (mine-006, mine-008). Later records settle what this fact first left open: the board memo gives Section 16 as the lease (csite-002), and the board approved the arrangement on September 17, 2026 (mine-011, csite-001). The notice named neither Valar nor a nuclear use (csite-005, mine-012).

In plain termsIn the public state records found, Valar's name is first tied to its Carbon County site in August 2026, as a lease of a square mile of state trust land north of Wellington. The state board approved it on September 17, 2026 (csite-001).
verified 2026-10-15: Second check, late September 2026: agenda PDF (item 10A, SULA 2095 OBA, 640 acres, Valar Atomics) and Trust Lands notice (Sec. 16 T14S R11E, near Wellington, bids by Aug 7, applicant unnamed; page now marked 'archived') re-read. Match; quotes exact.
analysisconfirmed 'In record time': the company's short timelines start in late 2025; some of its reactor hardware was built earlier
2026-08-04 · orig-060 · Valar Atomics, Ward 250, Ward Zero

Claims: the technology page says Valar designed and built its test reactor 'in record time'; the Series B post says it took seven months to take Ward 250 critical; Taylor, quoted in DOE's release, said the site was empty nine months before criticality (orig-010); the June 2026 draft counts 'only 9 months' from groundbreaking to power. Record: Valar's own safety and QA papers say Ward 250 incorporates Ward Zero hardware (orig-059), and Ward Zero took about ten months to design and build, by Taylor's account to TechCrunch, before its February 2025 completion (orig-005), seven months before the Utah groundbreaking. The core physics was tested in Nevada in November 2025 on a Los Alamos machine (orig-008), and the Series B post itself says the NOVA core took two years. The short timelines hold as worded for the Utah site and for the step from the Nevada core test to Utah criticality (seven months, November 2025 to June 2026; orig-015), but the reactor was the product of roughly two years of earlier work in California and Nevada.

In plain termsValar's short timelines (seven and nine months) count from late 2025. By its own documents, the reactor reuses hardware components from a prototype built and tested in California and finished in February 2025, and its core physics was tested at a federal lab in November 2025. The short timeline is true for the Utah site, not for the reactor as a whole.
verified 2026-10-15: Second check, late September 2026: technology page, Series B post, TechCrunch and the draft page re-read. Removed the claim that 'seven months' fits no milestone: it matches NOVA criticality to Ward 250 criticality.
analysisconfirmed More Series B lines checked: 'a week later' was 13 days; 'no burden on our water supply'; Valar making its own fuel
2026-08-04 · orig-070 · Valar Atomics, NVIDIA

(1) The Series B post says that 'just a week later' than the June 18 criticality, Ward 250 generated electricity to power an NVIDIA Blackwell. Valar's own Castle Country page and Deseret News date that event July 1, 2026, thirteen days after criticality (orig-013, orig-042), and TechCrunch, attributing it to the company, wrote that it happened in June. A small point, but the company's own pages disagree. (2) The post says its power will be generated in reactors with no burden on the water supply. The Trust Lands memo says the helium-cooled reactors need no cooling water (mine-017), which supports the reactor part; separately Valar is seeking up to 1,000 water shares in Emery County for future development (mine-042) and would have to buy water rights for its Carbon County campus (mine-017). The claim holds for reactor cooling; the wider campus needs water. (3) The post says Valar will not rely on outside suppliers for fuel and will make it in labs beside its reactors. So far, DOE's shipping review cleared HALEU fuel for Ward 250 to be shipped from Los Alamos's criticality center in Nevada (safe-010); DOE's fuel-line selection of Valar is conditional (orig-032); and no DOE authorization of the Valarin fuel line was found in public records as of September 2026 (safe-031). A plan, not the current practice. (4) 'Ward 250's operability has been proven': DOE confirmed criticality (orig-010); Valar says it reached full power days after criticality, but no independent record of the power reached or of the run history was found in public records as of September 2026 (orig-013, safe-007).

In plain termsChecking more lines from Valar's big funding announcement: the electricity demonstration came about two weeks after start-up, not 'a week'; the reactors themselves need no cooling water, but the company is seeking water shares for its wider plans; and Valar's fuel so far was shipped from a federal laboratory site, not made in its own plant, which is still a plan.
verified 2026-10-15: Second check, late September 2026: Series B post, Castle Country page and TechCrunch ('In June ... the company said') re-read; quotes exact. DOE's shipping review, relied on for point 3, read and added.
company claimconfirmed The $1B plan in Valar's words: fleets 'en masse', then thousands a year; no site, model, factory or date is named
2026-08-04 · plan-020 · Valar Atomics, U.S. Department of Energy, U.S. Nuclear Regulatory Commission · Series B equity, as announced by Valar: $1,000,000,000

Valar's Series B post (4 August 2026) says the money lets it move from demonstrating one integrated reactor to 'producing fleets of them en masse', with each reactor shortening the next until Valar produces 'tens, then hundreds, then thousands of reactors per year' in factories 'filled wall to wall' with small modular reactors, and that it will make its own fuel beside its reactors. It names no next reactor design or size, no factory site, no customer and no date. The concrete signs of the next step are elsewhere: a manager for 'one of Valar's first commercial projects' in Utah on the job board (plan-011), a turbine program (plan-018), the Carbon lease (mine-011, plan-016), a state tax-credit application projecting $1.3 billion of investment and 275 jobs in Emery County (fund-017), and the 'Commercial Deployment by 2028' target in Valar's July 2025 Emery County slide deck (orig-045). DOE's July 2026 fact sheet, in its paragraph on Antares's Mark-0, says that demonstration set a baseline for 'subsequent reactors' to produce electricity 'in 2027 and the following years'; it gives no electricity date for Valar. Executive Order 14301's finding that test reactors fall under DOE covers reactors that do 'not produce commercial electric power' (safe-030, csite-010), and federal law gives the NRC licensing authority over demonstration reactors 'operated as part of the power generation facilities of an electric utility system' (42 U.S.C. 5842); no NRC application or pre-application from Valar was found in public records as of late September 2026 (safe-032).

In plain termsValar raised a billion dollars on the promise of mass-producing reactors, eventually thousands a year. The announcement does not say what the next reactor is, where it will be built, or when. The order that put Ward 250 under DOE covers reactors that do not produce commercial power; a reactor selling power would fall outside it, and no application from Valar to the Nuclear Regulatory Commission was found in public records as of late September 2026.
verified 2026-10-02: Checked in late September 2026: Series B post, DOE fact sheet and the July 2025 deck (1295975.pdf, 12 pages) re-read; quotes exact. Fixed: the fact sheet's 2027 sentence sits in its Antares paragraph and was described too broadly. Corrected in V1.1 (October 2026): the NRC licensing line follows Executive Order 14301 and 42 U.S.C. 5842; see Corrections.
company claimconfirmed Taylor's 'first company ever' to take a reactor critical outside a national lab: DOE said something narrower
2026-08-04 · ppl-030 · Isaiah Taylor, Valar Atomics, U.S. Department of Energy, General Atomics

In the Series B post he signed (4 Aug 2026), Taylor wrote that Ward 250's criticality made Valar 'the first company ever to take a nuclear reactor critical outside of a national lab', and that Ward 250 was 'the first time in history a startup generated nuclear power'. DOE's own release (18 Jun 2026) makes a narrower claim: the first DOE-authorized reactor built outside a national laboratory. The broader wording does not hold as written: a conference history indexed by DOE's OSTI database records that the first TRIGA research reactor, designed by the private company General Atomics (founded in San Diego in 1955 by General Dynamics), was commissioned at General Atomics on 3 May 1958, and Valar's own safety agreement cites the privately run Vallecitos Nuclear Center as precedent (orig-039). The 'startup' electricity claim could not be tested; no independent record of the power produced was found in the public records searched as of late September 2026 (orig-013). Two timing claims by Taylor also differ in length, though they need not conflict: 'Nine months ago, this was an empty site' (DOE release, June 2026, consistent with the September 2025 groundbreaking, orig-007) and 'It took seven months to take Ward 250 critical' (Series B post), which gives no start date. Compare the earlier NOVA 'first' (orig-047), which The New Republic says Taylor rowed back.

In plain termsValar's chief executive wrote that his company was the first ever to start up a reactor outside a national laboratory. A private company, General Atomics, started one in San Diego in 1958. The Energy Department's own wording was narrower: the first reactor it has authorized outside its labs.
verified 2026-10-15: Re-read the Series B post, DOE's 18 Jun 2026 release and the OSTI TRIGA abstract; this time osti.gov answered a direct request, and the quote 'First TRIGA reactor Mk-1 was commissioned on 3 may 1958 at G.A.' was machine-checked verbatim. The abstract also states General Atomics was founded in 1955 in San Diego by General Dynamics. Conflict confirmed as written. Small fix: the two timing claims differ in length but need not conflict. Revised before publication (late September 2026): plain's closing line restated.
company claimreported Valar announces $1B Series B led by Sequoia plus a $200M credit facility
2026-08-03 · orig-014 · Valar Atomics, Sequoia Capital, Shaun Maguire, Erebor Bank, J.P. Morgan, Crescent Cove, Hercules Capital, Riot Ventures · Series B equity (company-announced): $1,000,000,000; credit facility (company-announced): $200,000,000

A post signed by Isaiah Taylor, dated August 4, 2026 in Valar's library, announces a closed $1 billion Series B led by Sequoia Capital, with Sequoia partner Shaun Maguire joining the board, and a separate $200 million credit facility led by Erebor Bank and J.P. Morgan with Crescent Cove and Hercules Capital. Named co-investors include Apandion, Atreides Management, Conviction, Dream Ventures, HOF Capital, Point72, Riot Ventures, Snowpoint Ventures and Valor Equity Partners. The post says the money will move Valar from one integrated reactor to fleet production, including making its own fuel in labs beside its reactors. No valuation is stated. The full funding history is in orig-057. TechCrunch reported the announcement on Monday, August 3, 2026, a day before the date the post carries, and gave Bloomberg's report of a $6 billion valuation.

In plain termsIn August 2026 Valar announced $1 billion in new investment led by Sequoia, plus a $200 million credit line. A private company does not have to publish its accounts, and as of October 2026 no terms beyond the announcement were found in public records (orig-057).
verified 2026-10-15: Second check, late September 2026: post re-read; round, board seat, co-investors, credit facility and fuel-lab plan match. Event date corrected to August 3 (TechCrunch reported the announcement that Monday; the post is dated August 4).
analysisreported Independent assessments, 2026: the airlift moved an unfueled, non-working unit in three pieces; criticality is not power
2026-07-09 · orig-073 · Valar Atomics, Partnership for Global Security, MIT Technology Review

Partnership for Global Security (its president, Ken Luongo, March 13, 2026) argues the C-17 flight was political and public-relations 'cinema'; that what flew was a single non-operational, unfueled unit delivered in three pieces; that the Wall Street Journal reported Valar paid for the flights, roughly estimated at under $1 million (WSJ not read); that U.S. military aircraft had carried small reactors before, including two Russian TOPAZ-II space reactors flown by C-5 in 1992; that Valar is not part of any military reactor project (the Defense Department's Project Pele reactor is being built by BWXT); and, citing former DOE nuclear chief Katy Huff, that a cold zero-power test does not show hot, full-power performance. It wrote that Valar's reactor 'seems to have some advantages' because the U.S. government had approved its preliminary safety analyses. It also wrote that no power production would occur at the Utah site; Valar says, and Deseret News reported, that Ward 250 made electricity there on July 1, 2026 (orig-013, orig-042). MIT Technology Review (July 9, 2026) reports that four pilot reactors, Valar's among them, reached zero-power criticality by the July 4 deadline; that criticality and running a reactor that can make electricity are two different things; that startup timelines deserve caution; and that a Third Way memo called the federal push an 'unhelpful diversion'. Both are analysis, not findings by a regulator.

In plain termsTwo independent assessments put Valar's big moments in context. The airlift carried an unfueled, non-working unit, and the military had flown reactors before. Reaching criticality is a first step, not proof that a reactor can make power reliably. MIT Technology Review calls the speed impressive; both say the hard parts are still ahead.
verified 2026-10-15: Second check, late September 2026: PGS (cinema, unfueled three-piece unit, WSJ report that Valar paid under $1M, TOPAZ-II by C-5 in 1992, Pele by BWXT, Huff) and MIT Technology Review (four reactors, criticality vs electricity, timelines, Third Way) re-read; quotes exact. Plain no longer credits PGS with praising the speed; the July 1 claim is attributed.
company claimreported Company: Ward 250 reached full power days after criticality and made first electricity on July 1, 2026
2026-07-01 · orig-013 · Valar Atomics, NVIDIA

Valar's Castle Country page says Ward 250 reached full power days after its June 18 criticality and generated electricity on July 1, 2026, live before an audience, directly powering an NVIDIA chip; its Series B post says an NVIDIA Blackwell was powered and calls this the first time an advanced reactor powered AI infrastructure. The same page calls Ward 250 a research reactor, not a commercial power plant, about the footprint of a shipping container, and says a June 26 test shut it down and cut active cooling and power to show passive decay-heat removal. Deseret News independently reported the July 1 event itself (orig-042). A DOE slide prepared for the NRC Commission's 21 July 2026 meeting lists 'July 2, 2026 – full power 100 kWt full power achieved', with electricity generation for an NVIDIA computer in the same line, a day after Valar's date for the electricity; the slide gives no data or source for that line. DOE's Theodore Garrish told the Commission that during the initial criticality test Valar 'demonstrated that they can produce electricity and powered a chip'. No measured power record and no independent record of the June 26 test were found in public records as of 17 October 2026 (safe-007).

In plain termsValar says the reactor reached full power a few days after June 18 and made its first electricity on July 1, which it says directly powered an NVIDIA chip at a public event. A DOE slide shown to federal regulators in July lists full power on 2 July, without data; no measurement of the power level and no DOE operating record were found in public records as of 17 October 2026 (safe-007).
verified 2026-10-17: Second check, late September 2026: Castle Country page and Series B post re-read; all company statements present. Plain text reworded so it does not imply a measured power level.
recordreported July 1, 2026 Orangeville event: Valar and NVIDIA announce they are exploring a 30 MW nuclear-powered data center
2026-07-01 · orig-042 · Valar Atomics, NVIDIA, Isaiah Taylor, Emy Lesofski, David Hinkins, Deseret News

Deseret News (Eva Terry, July 1, 2026) reported an event at Valar's Orangeville site with Taylor, an NVIDIA energy-policy director, Utah energy office director Emy Lesofski and State Sen. David Hinkins, where the companies said they are exploring a 30 megawatt data center fully powered by Valar's reactors. The article says Ward 250 was 'now generating 100 kilowatts of electricity'; that wording cannot match the 100 kW continuous thermal limit in Valar's draft safety agreement (orig-034), which allows 250 kW only for up to an hour: a reactor limited to 100 kW of heat cannot deliver 100 kW of electricity, because turning heat into electricity always loses part of it. DOE's April 2026 review calls Ward 250 a 250 kW thermal reactor; even at 250 kW of heat, 100 kW of electricity would mean turning 40% of the heat into electricity. Valar's Castle Country page says the reactor generated its first electricity at the July 1 event, without giving an amount (disc-044). The figure is most likely a thermal figure reported as electric.

In plain termsOn July 1, 2026 at the Emery County site, Valar and NVIDIA said they are exploring a 30-megawatt data center run on Valar reactors. The newspaper's '100 kilowatts of electricity' cannot be right for a reactor limited to 100 kilowatts of heat in continuous running, as Valar's draft safety paper sets, because only part of the heat can become electricity; DOE describes the reactor as 250 kilowatts of heat, and even then 100 kilowatts of electricity would take 40% of the heat.
verified 2026-10-02: Second check, late September 2026: Deseret re-read: event, speakers (an NVIDIA energy-policy director, Lesofski, Hinkins), 30 MW data-center study and '100 kilowatts of electricity' all present. The inference is now stated with its reason. Corrected in V1.1 (October 2026): adds DOE's 250 kW thermal figure; see Corrections.
company claimconfirmed On an unlisted draft page, Valar calls itself 'very open and transparent', pointing to its photos and video; of its Ward 250 safety papers, two were found published, and the others, most listed for DOE review, were not found in public records
2026-06-28 · safe-040 · Valar Atomics

In its unlisted 'Ward 250 By the Numbers' page (28 June 2026), Valar says its beliefs have led it to be 'a very open and transparent company', that its reactor has been more photographed and filmed than any other in history, and that it is sharing raw data to be refined by internal and external peer review over the following weeks. The page still contains placeholders ('July XX', '[xxx]') and one chart, 'Reactor Power Ascension', which plots an unlabeled 'Q_Total' on a 0-100 scale and thermocouple readings of about 30-80 F against a 'sample index', with no power units, dates or time scale, so it cannot be used to check the power reached. Of the safety documents the NSDA and DOE records name for Ward 250, two were found published by Valar: the NSDA (marked 'Draft', safe-005) and the QA Program Description (safe-018). Not found among Valar's published documents as of late September 2026: the PDSA, final DSA and Technical Safety Requirements, the Code of Record (document 100437), the startup and commissioning plan, the radiation protection program, the transport QA program (document 100719, which DOE approved, safe-012) and any emergency plan. The Physical Security Plan and Material Control and Accountability Plan are security documents, which the NSDA calls separate controlled documents. County and state officials praised Valar's openness at the Trust Lands board meetings (csite-028).

In plain termsValar describes itself as unusually open, and it has published many photos and videos. The safety documents are a different matter: of the documents that define how the reactor is kept safe, we found two the public can read, and one is labeled a draft, though Valar says DOE approved it; the others were prepared for DOE's review. Some documents, like security plans, are normally kept private. The safety analysis and operating limits are not security plans; they can be requested from DOE under FOIA (safe-007).
verified 2026-10-02: Re-read 'By the Numbers' (single embedded image, the 'Reactor Power Ascension' chart, read from the image: Q_Total 0-100 with no units, thermocouples about 30-80 F, x-axis 'Sample index'), NSDA and QA Approval 0049 in late September 2026. Holds. Softened 'has published two' to 'two were found published': Valar's docs library page as posted lists five posts and no safety documents, so a complete inventory could not be made. Corrected 2026-09-30: reworded the plain-language text.
recordreported ANS: El Segundo startup; Ward 250 is 100 kWt per Taylor; DOE Idaho office ran the start-up review
2026-06-22 · orig-011 · Valar Atomics, American Nuclear Society, Isaiah Taylor, Bob Boston, DOE Idaho Operations Office

ANS Nuclear Newswire (June 22, 2026) identifies Valar as an El Segundo, California-based start-up and reports that Isaiah Taylor has indicated Ward 250 is a 100-kWt (thermal) reactor built to demonstrate its TRISO-fueled HTGR technology; the power figure is attributed to Taylor, not independently measured. It reports Valar announced 10 kWt output on June 22 during power ascension. It says the federal review was led by Bob Boston, a manager at DOE's Idaho Operations Office, and that a joint test group had to sign off on key start-up steps before DOE authorized the criticality test. It also notes the executive order's three-reactor-by-July-4 target was still unmet at that date.

In plain termsA nuclear trade outlet reported that Ward 250 is rated at 100 kilowatts of heat, a very small reactor (about the heat output of three or four large home furnaces), and that a DOE Idaho manager led the federal review. The power rating comes from the company.
verified 2026-10-15: Second check, late September 2026: ANS re-read: El Segundo, 100 kWt per Taylor, 10 kWt on June 22, Bob Boston of DOE Idaho, joint test group, three-reactor target unmet. All match; quote exact.
recordconfirmed DOE: Ward 250 completed zero-power fueled criticality in Emery County, second under the Reactor Pilot Program
2026-06-18 · orig-010 · Valar Atomics, U.S. Department of Energy, Chris Wright, Isaiah Taylor, Antares Nuclear

DOE's June 18, 2026 release says Ward 250 completed a zero-power fueled criticality demonstration at the Utah San Rafael Energy Lab in Emery County, the first DOE-authorized reactor built outside a national laboratory. DOE calls it the second of several reactors expected to go critical by the July 4 deadline in the May 2025 executive order; Antares Nuclear's Mark-0 at Idaho National Laboratory was first earlier that month. Energy Secretary Chris Wright and Isaiah Taylor are quoted; Taylor says the site was empty nine months earlier. DOE describes criticality as a sustained chain reaction that must precede power generation.

In plain termsDOE confirms that on June 18, 2026 Ward 250 in Emery County sustained a chain reaction at essentially zero power, the first step before a reactor makes heat. It was the second reactor to do this under DOE's pilot program. DOE says criticality must be achieved before a reactor can generate power; it shows a sustained chain reaction, not how the reactor performs at power, and it does not settle the separate questions about its safety documents (exp-005, safe-007).
verified 2026-10-02: Second check, late September 2026: DOE release and ANS re-read: zero-power fueled criticality, first DOE-authorized reactor built outside a national lab, second after Antares, quotes exact. Revised before publication (2026-10-01): wording made more exact against the cited records.
recordconfirmed POWER and a DOE slide: DOE gave preliminary safety-analysis approval in Feb 2026 and final approval April 23, 2026
2026-04-23 · orig-033 · Valar Atomics, U.S. Department of Energy, POWER magazine

POWER magazine's June 19, 2026 report says Ward 250's DOE Documented Safety Analysis received preliminary approval in February 2026 and final approval on April 23, 2026, ahead of the June 18 criticality. It describes Ward 250 as a TRISO-fueled modular HTGR with helium coolant and names applications including electricity, process heat, hydrogen and materials processing. A DOE Office of Nuclear Energy slide prepared for the NRC Commission's 21 July 2026 meeting lists the same dates, 'February 2026 – PDSA Approved' and 'April 23, 2026 - DSA Approved', with no document numbers. The DSA itself was not read: no public copy was found as of 17 October 2026 (safe-007). POWER also names Kiewit Nuclear Solutions as the engineering, procurement and construction contractor.

In plain termsA trade magazine reported that DOE approved Ward 250's safety analysis in two steps, in February and April 2026, and a DOE slide shown to federal regulators in July 2026 lists the same two dates. No public copy of the safety analysis itself was found (safe-007).
verified 2026-10-17: Second check, late September 2026: POWER re-read: preliminary DSA approval Feb 2026, final April 23, applications list. Match. Raised to confirmed on 17 October 2026: a DOE slide prepared for the NRC Commission's 21 July 2026 meeting lists the same two approval dates; no public copy of the DSA was found. Corrected 1 Oct 2026: wording made more exact against the cited records.
recordconfirmed USPTO record: Valar Atomics Inc. is a Delaware corporation; it applied in March 2026 to trademark its name
2026-03-03 · orig-051 · Valar Atomics, U.S. Patent and Trademark Office

The U.S. Patent and Trademark Office's file for application serial 99680835 (read in late September 2026) lists the applicant as Valar Atomics Inc., a corporation organized in Delaware, at 4857 W 147th Street, Hawthorne, California, represented by the law firm Goodwin Procter. The application, filed March 3, 2026 for the word mark VALAR ATOMICS, covers 'nuclear reactors' (class 11) and design and engineering services for nuclear reactors and for 'gigasites' (class 42). A voluntary amendment was received on April 30, 2026; on July 9, 2026 the examining attorney issued a non-final office action (a letter asking for more information or making an initial refusal), and on September 11, 2026 an application extension was granted, so the mark was not registered as of late September 2026. Whether Valar has filed for any other marks, such as its product names, could not be checked as of late September 2026: the USPTO trademark search site could not be read. This is the first readable government record of Valar's state of incorporation; Delaware's own registry sits behind a CAPTCHA and was not read, so the date of incorporation remains unconfirmed (orig-053).

In plain termsA federal trademark filing made by the company's own lawyers says Valar Atomics Inc. is a Delaware corporation headquartered in Hawthorne, California. Many U.S. startups incorporate in Delaware for its business law. The filing also shows how Valar describes its business to the government: reactors and 'gigasites'.
verified 2026-10-15: Second check, late September 2026: TSDR re-read: registrant, Delaware, Hawthorne address, Goodwin Procter, classes 11 and 42 (quote exact), filing and office-action dates. The trademark-search source was removed: tmsearch.uspto.gov now serves a bot challenge, so the 'no other filings' result could not be re-checked and was taken out of the detail.
recordconfirmed Feb 2026 Emery County agenda: support letter for Valar and an MOU for site testing on a county parcel
2026-02-17 · orig-046 · Valar Atomics, Emery County Commission

The Emery County Commission's February 17, 2026 meeting notice lists two Valar items: approving or denying a county letter of support to Valar Atomics, and approving or denying an MOU letting Valar perform site suitability testing on parcel 04-0019-0030. This shows Valar was seeking to test county land next to the state lab by early 2026: county minutes of 20 January 2026 call the parcel 'the property next to the Research Center', and the county later approved selling it to Valar (gov-022). Valar's Castle Country page says the commission backed the project unanimously. The commission's minutes of the same day record the support letter approved 3-0, subject to the county attorney's review and approval, and the MOU item as 'Tabled'.

In plain termsIn February 2026 Emery County's agenda listed a support letter for Valar and a proposed agreement letting Valar test a county parcel next to the state lab; the commission approved the letter and tabled the agreement. The county later approved selling that parcel, about 112 acres, to Valar (gov-022).
verified 2026-10-19: Second check, late September 2026: Utah Public Notice page re-read: February 17, 2026 agenda with the support letter and the parcel 04-0019-0030 MOU. Castle Country page still says the commission backed the project unanimously. Match.
recordconfirmed Ward 250 flown unfueled from March ARB, Calif., to Hill AFB, Utah, on three C-17s (Operation Windlord)
2026-02-15 · orig-030 · Valar Atomics, U.S. Department of Energy, U.S. Department of War, U.S. Air Force, Chris Wright, Michael Duffey

World Nuclear News (Feb 18, 2026) reports the Ward 250 reactor, without its nuclear fuel, was loaded onto a C-17 at March Air Reserve Base in California on February 15, 2026 (Stars and Stripes reports three C-17s carried the modules) and flown to Hill Air Force Base, with Energy Secretary Chris Wright and Under Secretary Michael Duffey aboard, for later transfer to the Utah San Rafael Energy Lab. WNN, citing Reuters, says Wright stated fuel would move separately from the Nevada National Security Site. ExecutiveGov describes the same flight as a DOW-DOE partnership. Both call Ward 250 a 5-megawatt reactor; Valar's safety agreement sets its maximum continuous power at 100 kilowatts of heat (orig-042). This shows the reactor hardware was in Southern California before it moved to Utah; Valar's June 2026 draft says it was tested at its Hawthorne headquarters (orig-059). See orig-027 for the dispute over whether it was a complete reactor.

In plain termsIn February 2026 three Air Force cargo jets flew the reactor's modules, without fuel, from Southern California to Hill Air Force Base, and trucks took them on to Emery County.
verified 2026-10-02: Second check, late September 2026: WNN and ExecutiveGov re-read: unfueled, March ARB, Feb 15, Hill AFB, Wright and Duffey aboard, fuel to move separately (Reuters via WNN). One overstated inference softened.
analysisreported Neutron Bytes calls the Feb 2026 C-17 cargo a 'mock up'; Valar says Ward 250 itself flew in modules
2026-02-15 · orig-027 · Valar Atomics, U.S. Department of Energy, U.S. Department of War, Neutron Bytes, Hill Air Force Base

Valar's Castle Country page says Ward 250 flew to Utah in modules aboard three C-17s on February 15, 2026 ('Operation Windlord'), with DOE and the Department of War, and then went by road from Hill Air Force Base to Emery County. Neutron Bytes captions the cargo as components of a 'mock up' of the Valar microreactor and argues a real transport test would need a complete, shielded reactor. Neither source read says whether fuel was aboard. World Nuclear News reports the Ward 250 reactor was flown without its fuel, and Stars and Stripes says the reactor's modules were loaded (orig-030); no other record read calls the cargo a mock-up. Other reports on the airlift: gov-027.

In plain termsValar calls the February 2026 flight the first-ever C-17 airlift of a nuclear reactor. A critic called the cargo parts of a mock-up. Federal and trade reports say the reactor flew without fuel, which was to follow separately from Nevada (orig-030, orig-073).
verified 2026-10-15: Second check, late September 2026: Castle Country page and Neutron Bytes re-read; 'mock up' caption and 'real test' sentence present; quotes exact.
analysisconfirmed Zero-power criticality shows little about a working reactor, says ex-DOE nuclear chief Katy Huff; Valar's own release draws the same line
2026-01 · orig-026 · Katy Huff, Neutron Bytes, Latitude Media, Valar Atomics

Katy (Kathryn) Huff, former DOE Assistant Secretary for Nuclear Energy, said a zero-power criticality test 'can be achieved without making real engineering progress on fuel or design', since it only shows neutrons can multiply. Neutron Bytes (Dan Yurman, Feb 28, 2026) quotes it as said to Latitude Media 'last month' and names Valar's Los Alamos test as her example; MIT Technology Review (July 9, 2026) quotes the same words from the Catalyst podcast earlier that year. The original interview was not reviewed for this record. It is directed at claims built on the November 2025 NOVA result (orig-008). Valar's own NOVA release concedes the distinction ('Cold proves the physics. Hot proves the power.').

In plain termsA former head of DOE's nuclear office said a 'zero-power' test only shows the chain reaction works; it says little about whether a working power reactor will. Valar's own release draws a similar line: 'Cold proves the physics. Hot proves the power.'
verified 2026-10-15: Second check, late September 2026: Neutron Bytes and NOVA release re-read. MIT Technology Review quotes Huff's same words, so a second independent report was added and confidence raised; date set to January 2026 ('last month' per Neutron Bytes). Revised before publication (2026-10-01): wording made more exact against the cited records.
recordconfirmed Nov 2025: $130M Series A reported, led by Snowpoint Ventures, with Palmer Luckey among backers
2025-11 · orig-049 · Valar Atomics, Snowpoint Ventures, Doug Philippone, Palmer Luckey, Shyam Sankar, Dream Ventures, Day One Ventures, Contrary · Series A (reported): $130,000,000

Contrary Research (an investor) reports a $130 million Series A in November 2025 led by Doug Philippone of Snowpoint Ventures, with Palmer Luckey, Palantir CTO Shyam Sankar, Dream Ventures, Day One Ventures and Contrary participating, bringing total funding to about $150M then. Valar's press library links a Bloomberg headline naming Luckey and a Lockheed director as backers, and a later Los Angeles Times B2B headline 'El Segundo's Valar Atomics Raises $130 Million'; neither article could be read. Full round-by-round detail is on the Funding tab. ANS (November 13, 2025), a second, independent report, gives the same $130 million and says the round was led by Snowpoint, Day One and Dream, with Snowpoint's cofounder joining the board and total funding above $150 million.

In plain termsIn November 2025 Valar raised $130 million, led by Snowpoint Ventures, with defense-technology figures among its backers.
verified 2026-10-15: Second check, late September 2026: Contrary re-read (lead, backers, ~$150M total). ANS's independent report of the same round added, so confidence raised to confirmed; the two differ on which firms 'led'.
company claimconfirmed NOVA release gives the design basis: HALEU TRISO fuel, graphite moderator, helium loop, B4C control
2025-11-18 · orig-009 · Valar Atomics, Los Alamos National Laboratory, Sonat Sen

The NOVA release is the most specific company-published technical description found. It says NOVA used the same fuel, moderator and reactivity-control scheme as Ward 250 (HALEU TRISO fuel, graphite moderator, boron-carbide control elements in stainless steel), and that its data would inform helium-loop operations and temperature ramp-up in Valar's reactor program. It credits the core test geometry to LANL's earlier Deimos critical assembly (2024), and quotes Valar's lead core designer, Sonat Sen, saying the data would validate Valar's proprietary design software. DOE defines HALEU as uranium enriched between 5% and less than 20% U-235 (today's reactors use fuel up to 5%); the release does not give Valar's exact enrichment.

In plain termsThis release is the most detailed public description of Valar's core: coated-bead fuel of the richer HALEU grade, graphite, and boron control pieces. It says the Nevada test core used the same fuel as the Utah reactor, which matters for the fuel-grade conflict in disc-001.
verified 2026-10-15: Second check, late September 2026: release re-read; all design points match. The HALEU definition had no cited source: DOE's own definition page added.
analysisconfirmed 'First startup in history to split the atom': the NOVA 'first' ran on a Los Alamos machine under LANL operation
2025-11-18 · orig-047 · Valar Atomics, Los Alamos National Laboratory, Gizmodo

Valar describes Project NOVA as making it the first startup, or first venture-backed company, to achieve criticality. The joint release itself says LANL operated the core on its own Comet critical assembly, supplying the facility safety envelope, reflectors, instruments and analysis, while Valar built the central core and supplied fuel. Gizmodo's headline framed it as a claim ('California Startup Claims Historic First') and stressed that reaching criticality does not mean a reactor is ready to generate power. The New Republic (May 26, 2026) reported that Valar's founder, Isaiah Taylor, later backed away from the first-startup claim; Valar's Castle Country web page still made it when read in late September 2026. The later, stronger claim, the first DOE-authorized reactor built outside a national laboratory (June 2026), is DOE's own wording (orig-010).

In plain termsValar calls itself the first startup to 'split the atom'. The test ran on a Los Alamos laboratory machine, operated by laboratory staff, with Valar's core and fuel inside. The New Republic reported in May 2026 that Valar's founder later backed away from the claim; Valar's Castle Country web page still made it in September 2026.
verified 2026-10-15: Second check, late September 2026: NOVA release (LANL operated the Comet assembly; Valar built the core and supplied fuel; 'venture-backed' first) and Gizmodo (Nov 18, 2025; 'Claims'; criticality is not power) re-read. Match; quotes exact.
recordconfirmed Project NOVA: Valar-built HALEU TRISO core reached zero-power criticality at LANL's NCERC in Nevada
2025-11-17 · orig-008 · Valar Atomics, Los Alamos National Laboratory, National Criticality Experiments Research Center, National Nuclear Security Administration, Nevada National Security Site

A joint Los Alamos National Laboratory and Valar announcement (posted on Valar's site, dated November 18, 2025) says the NOVA core reached zero-power ('cold') criticality at 11:45 a.m. PT on November 17, 2025, at LANL's National Criticality Experiments Research Center on the Nevada National Security Site, after approach-to-criticality work began November 12. Valar built the central core and supplied the TRISO fuel; LANL operated it on NCERC's Comet critical assembly and supplied the facility safety envelope, instruments, reflectors and analysis, under NNSA Nevada Field Office oversight. The core is graphite-moderated, HALEU TRISO-fueled, with boron-carbide control elements, and was configured to model the Ward 250 core. The release itself stresses there was no power production, grid connection or full-temperature operation. The claim of the first criticality by a venture-backed company is the company's. POWER magazine (June 2026) independently describes the same November 2025 campaign and notes it ran under NNSA oversight at LANL, not under the Reactor Pilot Program authorization.

In plain termsIn November 2025 a core built by Valar and loaded with its fuel reached a self-sustaining chain reaction inside a Los Alamos laboratory machine in Nevada, run by Los Alamos staff. 'Zero power' means the reaction made no useful heat or electricity; it tests the physics only.
verified 2026-10-02: Second check, late September 2026: release re-read: times, dates, roles (Valar core and fuel; LANL assembly, safety envelope, reflectors, instruments, analysis), NNSA Nevada Field Office oversight and the no-power caveat all match. It is a company-hosted joint release that quotes a DOE official; no separate LANL posting was read. POWER's independent account added as a second source.
analysisreported Even an investor-published profile lists major unproven risks: mass reactor production, S-I hydrogen, synfuel costs
2025-11-11 · orig-024 · Valar Atomics, Contrary Research

Contrary Research, itself an investor, lists key risks: no company has mass-produced reactors at the scale Valar plans; hundreds of reactors at one site has never been attempted, so one design flaw could hit a whole gigasite; the sulfur-iodine hydrogen cycle has never been deployed commercially; carbon-capture costs of roughly $100-600 per ton could undercut synthetic-fuel economics; and TRISO fuel has limited commercial operating history and potential supply constraints. It also notes HTGR technology has not been commercially deployed at scale. These are the report's own assessments.

In plain termsEven a research firm that invested in Valar lists big unproven steps in the plan: mass-producing reactors, running hundreds of them together, and making hydrogen and fuels in ways no one has yet done commercially.
verified 2026-10-15: Second check, late September 2026: Contrary risk section re-read; each listed risk is in the text; quote exact.
recordconfirmed Ward 250 safety agreement: 100 kWth continuous (250 kWth brief tests), 4.95% LEU TRISO, 12-month operational life
2025-10 · orig-034 · Valar Atomics, U.S. Department of Energy, Ward 250

Valar's Nuclear Safety Design Agreement for Ward 250 (document 100403 rev 02, October 2025, marked 'Draft' and 'issued for Public Release'; the company says DOE approved it) lists key design parameters: 100 kWth continuous as the license limit, up to 250 kWth for limited tests; TRISO-coated UCO fuel qualified on the AGR program basis; helium coolant at 3.0 MPa; 250 C inlet and 650 C nominal core outlet; a 1.8 m by 1.5 m core of 76 fuel elements holding about 250 kg of heavy metal; 4.95% U-235 enrichment; graphite reflector. It sets a 12-month operating period and an 18-month lifecycle ending in decommissioning and return of the site to unrestricted use (DOE's April 2026 categorical exclusion describes a 24-month program: about 12 months of operation, 6 of cooldown and 6 of decommissioning; disc-007), with DOE authorization only and no planned NRC licensing. At a July 2025 Emery County town hall the name was explained as a reference to the nation's 250th birthday on July 4, 2026 (orig-045); the 250 kWth test ceiling is a separate figure.

In plain termsThis is Valar's own summary of the safety design it says DOE approved (the public copy is marked 'Draft'). It sets the power limit (100 kilowatts of heat, briefly up to 250), the fuel (listed as 4.95% enriched uranium) and a 12-month run followed by cleanup. It lists the fuel differently from DOE's later shipping review: see disc-001.
verified 2026-10-15: Second check, late September 2026: NSDA PDF re-read (Table 6, p.52; executive summary; lifecycle section): every parameter matches, including DOE-only authorization and return to unrestricted use. The web post says DOE approved it. Quote exact. Revised before publication (late September 2026): plain: DOE's approval is Valar's statement and the copy is marked 'Draft' (detail already said both).
company claimconfirmed QA plan covers a 'Valarin' TRISO fuel fabrication facility at the Utah lab, based on German HOBEG methods
2025-10 · orig-038 · Valar Atomics, Valarin Fuel Fabrication Facility, HOBEG

Valar's QA Program Description (October 2025; status 'Approved', and Valar's post says DOE approved it) covers not just the Ward 250 reactor but a 'Valarin Fuel Fabrication Facility' at the Utah San Rafael Energy Research Center; together they are called the Valar Atomics San Rafael (VASR) Facility. It says the fuel lab will make TRISO particles in graphite compacts using German HOBEG technology with modern improvements. This matches Valar's stated aim to make its own fuel beside its reactors (orig-014) and its fuel-line pilot selection (orig-032). Whether the fuel lab has been built was not stated in the records read; no DOE authorization for it was found in public records as of late September 2026 (safe-031).

In plain termsValar's quality plan covers not only the reactor but a small plant at the same lab to make its own coated-bead fuel, using methods first developed in Germany. No DOE approval to run that fuel line was found in public records as of late September 2026 (safe-031).
verified 2026-10-15: Second check, late September 2026: QA PDF re-read: Valarin Fuel Fabrication Facility, VASR, HOBEG with modern improvements. Match. 'DOE-approved' now attributed to Valar.
analysisconfirmed Ward 250's safety papers lean on Ward Zero tests and reuse its hardware; no Ward Zero test data was found in public records as of late September 2026
2025-10 · orig-059 · Valar Atomics, U.S. Department of Energy, Ward Zero, Ward 250

Valar's NSDA (October 2025) says Ward 250 directly uses hardware that has undergone full-scale testing, that design work focuses on nuclear-specific modifications to the proven WardZero configuration, and it credits Ward Zero testing for key safety points: natural circulation during a loss of coolant flow, and the capacity of the Reactor Cavity Cooling System. It also asks DOE to accept the final safety analysis at 90% design completion because the WardZero program has completed design, installation and testing of major safety-significant systems. Valar's QA Program Description, which Valar's October 13, 2025 post says DOE approved, has a section on using existing WardZero hardware: Ward Zero components may go into Ward 250 if they stay traceable and are evaluated, and Ward Zero operating data may count as design verification by comparison to 'similar proven systems'. Valar's Ward 250 page dates Ward Zero's completion to 15 February 2025, before that QA program was approved. Valar's June 2026 draft says the reactor underwent non-nuclear testing at its Hawthorne headquarters above 750 C and 6 MPa before shipment. No Ward Zero test report, data set or DOE review of it was found in public records as of late September 2026, and the page titled 'Ward Zero Whitepaper' holds no Ward Zero data (orig-058). Whether DOE examined the Ward Zero data would be shown by the PDSA and DSA, of which no public copy was found (safe-007).

In plain termsWard 250 was not built from scratch in Utah. By Valar's own safety papers it reuses parts of Ward Zero, the electric-heated prototype tested in California, and Ward Zero's tests are offered as proof that key safety features work, such as the reactor cooling itself by natural air flow. Some of that hardware was built before the quality program was approved by DOE, as Valar says it was; the program allows that if the parts are traced and checked. No Ward Zero test results were found in public records, so the public cannot check the evidence this part of the safety case rests on.
verified 2026-10-02: Second check, late September 2026: NSDA (full-scale tested hardware; proven WardZero configuration; natural circulation and RCCS credited to WardZero; DSA at 90% design because WardZero tested major safety systems), QA plan (WardZero hardware section; 'similar proven systems'), QA post and draft page re-read. Match; approval wording attributed to Valar's post. Revised before publication (late September 2026): whitepaper line now 'holds no Ward Zero data'; source title parenthetical now 'not listed in Valar's library'. Revised before publication (2026-09-30): reworded the plain-language text.
recordconfirmed Who approves Ward 250: DOE alone. Its safety agreement says no move to an NRC license is planned
2025-10 · safe-001 · Valar Atomics, U.S. Department of Energy, Nuclear Regulatory Commission

Valar's Nuclear Safety Design Agreement (NSDA, document 100403 rev 02, October 2025) says Ward 250 is a contractor-owned, contractor-operated facility 'under DOE jurisdiction', authorized under the federal nuclear safety rule 10 CFR 830, with DOE-STD-1271-2025 as its primary authorization standard, and that no transition to NRC licensing is planned or intended. DOE's NEPA determination of 22 April 2026 says the same: the facility 'will operate under U.S. Department of Energy (DOE) authorization'. The legal basis is Executive Order 14301 (23 May 2025): section 3 finds that advanced reactors under enough DOE control that are not built 'to produce commercial electric power' fall within DOE's jurisdiction rather than the NRC's, and section 5 orders a pilot program outside the national laboratories under 42 U.S.C. 2140, approving at least three reactors 'with the goal of achieving criticality' in each by 4 July 2026. DOE-STD-1271 itself exempts anything the NRC regulates. The text of DOE-STD-1271 contains no public-hearing or public-comment step (searched for 'hearing', 'public comment', 'comment period'); its foreword only invites emailed suggestions for improving the standard itself.

In plain termsTwo federal agencies can approve a reactor. The Nuclear Regulatory Commission (NRC) is the independent safety regulator for commercial reactors, and its licensing includes published notices and a chance for the public to ask for a hearing. The Department of Energy (DOE) can instead approve test reactors it controls. Ward 250 took the DOE route, so the agency that sponsors the pilot program is also the one that judges its safety, and the DOE rulebook used has no step where the public comments or asks for a hearing. This is the route the 2025 executive order set up (critics dispute its legal footing, safe-003).
verified 2026-10-02: Re-read in late September 2026: NSDA pp. 2 and 47, CX p. 1, EO 14301 secs. 3 and 5, DOE-STD-1271 sec. 2.1.d and full-text search. Fixed: headline said 'will never move' where the NSDA says no NRC transition is 'planned or intended'; the EO sets criticality by 4 July 2026 as a goal, not a requirement; Federal Register page corrected to 90 FR 22591.
recordconfirmed Sept 2025: DOE's 'conditional' fuel-line pick of Valar, at its own cost; Ward 250's fuel was cleared to ship from a federal lab site
2025-09-30 · orig-032 · Valar Atomics, U.S. Department of Energy, Terrestrial Energy, TRISO-X, Oklo, Los Alamos National Laboratory, Standard Nuclear

DOE's September 30, 2025 release is the primary record. In the second round of conditional selections for its Fuel Line Pilot Program, DOE selected Oklo, Terrestrial Energy, TRISO-X and Valar Atomics Inc (Hawthorne, CA), Valar's project being to support TRISO fuel fabrication for the Ward 250 deployment and possibly other high-temperature gas reactors. Each company pays all costs of building, running and decommissioning its fuel facility and must arrange its own nuclear feedstock; companies may apply for HALEU through DOE's HALEU Availability Program. DOE had earlier selected Standard Nuclear to build TRISO fuel facilities. World Nuclear News and Mother Jones ('one of four companies', 'last September') report the same selection, and Valar's NOVA release and Series B post list it. Set beside the record: DOE's May 2026 shipping review authorized use of a shipping package for limited shipments of HALEU TRISO fuel compacts to Ward 250 from Los Alamos's criticality center in Nevada, where the NOVA test ran (safe-010); the NOVA release says Valar 'provided' the TRISO fuel, but no record read says who manufactured it; Valar's QA plan describes a 'Valarin' fuel lab at the Utah site (orig-038), for which no DOE authorization was found in public records as of late September 2026 (safe-031); and the Series B post says Valar will make its own fuel beside its reactors (orig-070).

In plain termsDOE gave Valar conditional permission to set up its own fuel-making line, at its own cost. So far, though, DOE cleared the fuel for Ward 250 to be shipped from a federal laboratory site in Nevada, and who made it was not found in public records. Making nuclear fuel is a separate operation with its own hazards, and no approval papers for Valar's fuel line were found in public records as of late September 2026.
verified 2026-10-15: Second check, late September 2026: combined with a duplicate record of the same DOE selection. DOE release, WNN, Mother Jones and NOVA release re-read; DOE's shipping review added for the fuel's origin. 'Where the NOVA test had used it' softened to 'where the NOVA test ran': the review does not say it is the NOVA fuel. Revised before publication (late September 2026): Mother Jones cited as 'Mother Jones (Kiera Butler), 26 Feb 2026' (narrow quote; headline not repeated).
company claimconfirmed Ward 250 groundbreaking: Valar's page says 17 Sep 2025; state records put the ceremony on 11 Sep 2025
2025-09-11 · orig-007 · Valar Atomics, Ward 250, Utah San Rafael Energy Lab

Valar's Ward 250 page lists 'Ward 250 Groundbreaking' on 09.17.2025, and its Castle Country page lists a local news item from September 2025 about breaking ground at the Utah San Rafael Energy Lab. DOE's June 18, 2026 release quotes Taylor saying the site was empty nine months before criticality, consistent with a September 2025 start. Local news publication of the groundbreaking release is covered in orig-067. State records settle the ceremony date: the Office of Energy Development's slide update to the Legislature's interim committee (15 October 2025) says Valar 'held groundbreaking' on September 11, 2025, Emery County's 2 September 2025 minutes say ground would be broken 'on September 11', and the county economic board's minutes schedule it for that Thursday (gov-015, gov-021). 17 September 2025 is the day ETV News printed the joint state-Valar release (loc-034), which is the date Valar's page shows (disc-040).

In plain termsConstruction at the Emery County lab began in September 2025, about nine months before the reactor first went critical in June 2026. State records put the groundbreaking ceremony on 11 September 2025; Valar's page shows 17 September, the day the announcement was published.
verified 2026-10-02: Second check, late September 2026: Ward 250 page date and DOE's quote of Taylor ('Nine months ago...') re-read; consistent. Further check, late September 2026: the event date had been taken from the company's page (17 Sept); gov-015 and gov-021 (state and county records read in late September 2026) put the ceremony on 11 Sept 2025, so the date field and headline now carry the recorded event date, with Valar's date kept as its claim.
company claimconfirmed Valar announces DOE selection to reach criticality by July 4, 2026 under Executive Order 14301
2025-08-12 · orig-006 · Valar Atomics, U.S. Department of Energy, Isaiah Taylor, Spencer Cox, Chris Wright

In a post dated August 12, 2025, signed by Isaiah Taylor, Valar said DOE had selected it to achieve criticality on American soil by July 4, 2026, pursuant to Executive Order 14301. The post thanks President Trump and Energy Secretary Wright, and Governor Cox and his administration for their partnership, names the test site as the San Rafael Energy Research Center in Emery County, and says testing there can proceed without multi-year licensing delays under the new framework. It describes the test reactor as a TRISO-fueled high-temperature proof-of-concept and claims operating temperatures above 750 C. The post frames the effort as a 'second Manhattan Project'. DOE's own selection announcement is covered in orig-041.

In plain termsIn August 2025 the Energy Department picked Valar as one of the companies allowed to try to start a test reactor by July 4, 2026, under a presidential order. Being picked meant permission to try under DOE's oversight, at the company's own cost, not federal money to build (orig-041).
verified 2026-10-02: Second check, late September 2026: post re-read; selection, EO 14301, site, TRISO proof-of-concept, >750 C and second-Manhattan-Project wording match. Thanks wording tightened to the text.
recordconfirmed DOE names Valar Atomics Inc. among 10 firms in its Reactor Pilot Program; companies bear all costs
2025-08-12 · orig-041 · Valar Atomics, U.S. Department of Energy, James P. Danly

DOE's August 12, 2025 release launched the Nuclear Reactor Pilot Program with 11 projects from 10 companies, including Valar Atomics Inc., aiming for at least three test reactors critical under DOE authorization by July 4, 2026, at sites outside the national laboratories. The release states each company is responsible for all costs of designing, building, operating and decommissioning its test reactor, so selection itself carried no federal construction funding. This is the primary record behind Valar's own announcement of the same day (orig-006).

In plain termsDOE's own announcement: ten companies, including Valar, were picked in August 2025 for a program to get test reactors running by July 4, 2026. Each company pays all of its own costs.
verified 2026-10-15: Second check, late September 2026: DOE release re-read: 11 projects, 10 companies incl. Valar Atomics Inc., at least three by July 4, 2026, companies bear all costs. Quote exact.
company claimconfirmed July 2025 Emery County town hall: Valar pitched a 12-month test, ~40 jobs, 'commercial development by 2028'
2025-07-08 · orig-045 · Valar Atomics, Emery County Commission, Utah San Rafael Energy Lab, Jaron Wallace, Jeremy Pearson

ETV News (Julie Johansen, July 9, 2025) reported the Emery County Commission's July 8, 2025 town hall in Castle Dale, requested by Valar after an MOU was signed and before its contract with the state lab was final. As reported, Valar said the company was about two years old, had a non-fueled test reactor near Los Angeles, aimed for July 4, 2026 (the nation's 250th birthday, hence 'Ward 250') with commercial development by 2028, expected about 40 employees for roughly 12 months of testing with local contractors for construction, and chose Utah because Carbon and Emery counties are the state's energy hub. A Valar slide deck posted on the state-run Utah Public Notice Website (file created July 8, 2025, server date 9 July 2025; 11 of its 12 pages are marked 'CONFIDENTIAL. DO NOT DISTRIBUTE.', yet the file is publicly posted; the public body and notice it is attached to were not identified) says the same: under 250 kW, 30 effective full-power days over 12 months in a temporary facility, commercial deployment by 2028.

In plain termsAt a July 2025 town hall Valar told Emery County residents about a small 12-month test with about 40 jobs, and a goal of commercial reactors by 2028. By 2026 the company was planning far larger sites (mine-019).
verified 2026-10-02: Second check, late September 2026: ETV report and the slide PDF (created Jul 8, 2025; 'CONFIDENTIAL'; <250 kW, 30 EFPD, 12 months, temporary facility; commercial deployment by 2028) re-read. Match. Corrected 1 Oct 2026: wording made more exact against the cited records.
company claimconfirmed Valar's policy stance: 'regulation follows innovation', DOE as testbed, 'dominance' over non-proliferation
2025-05-23 · orig-019 · Valar Atomics, Isaiah Taylor, U.S. Department of Energy, U.S. Nuclear Regulatory Commission

On the day of the May 23, 2025 nuclear executive orders, Taylor published a long essay welcoming them. It argues that US policy should shift from a non-proliferation focus to 'dominance' (while calling non-proliferation something that 'will remain, an important goal'), that DOE should act as a reactor testbed rather than a second regulator, and that regulation should follow industry rather than precede it; it approvingly summarizes provisions such as abandoning the linear no-threshold (LNT) radiation model and bypassing the NRC for DOE pilot reactors. This establishes the company's public policy position; critics' views of those same orders are covered in the government and safety records. Its summary of the orders includes three terms not found in the signed texts: nine military bases (EO 14299 orders one Army reactor at a domestic base), DOE decisions within a one-year timeframe (EO 14301 speaks of operation within 2 years), and fuel approvals under six months (no such term). They may come from same-day White House fact sheets, which we could not read.

In plain termsValar's chief executive argues that nuclear rules should come after a technology is built and tested, not before, and that U.S. policy should aim at 'dominance'. That is the company's political position, and it explains why Valar favors DOE's pathway over the NRC's licensing.
verified 2026-10-15: Second check, late September 2026: essay re-read; dominance, DOE-as-testbed, regulation-follows-industry, LNT and NRC-bypass passages match; quote exact. Revised before publication (late September 2026): added the three summary terms not found in the signed orders.
company claimreported April 2025: Valar joins states' suit against the NRC over small-reactor licensing; names 'Ward One'
2025-04-07 · orig-016 · Valar Atomics, U.S. Nuclear Regulatory Commission, Last Energy, Deep Fission, State of Utah, State of Texas

In a post dated April 7, 2025, Isaiah Taylor announced Valar was joining Texas, Utah, Louisiana, Florida and Arizona (in the complaint, the Arizona State Legislature; dir-017, safe-028), and the reactor companies Last Energy and Deep Fission, in a lawsuit against the Nuclear Regulatory Commission. The argument, as Valar summarizes it, is that the Atomic Energy Act of 1954 narrowed federal licensing to reactors significant to defense or public safety, and that a 1956 rule wrongly swept in all reactors. The post says Valar's planned 'Ward One' test reactor is named in the suit. Litigation status and outcome are tracked in safe-028 and dir-017 to dir-022; ANS reported on 5 October 2026 that the states had since filed a notice of voluntary dismissal (the filing itself was not read; dir-022).

In plain termsIn April 2025 Valar joined Texas, Utah and other states in suing the federal nuclear regulator (the NRC), arguing small reactors should not need its license at all. Utah, which later hosted Valar's reactor, was one of the states in that suit; in October 2026 the American Nuclear Society's news service reported that the states had since filed to drop their claims (the filing itself was not read).
verified 2026-10-02: Second check, late September 2026: post re-read; co-plaintiffs, 1954 Act argument, 1956 rule and Ward One claim match; quote exact. The court docket was not read in this check; the case papers are covered in dir-017 to dir-022.
company claimreported Design lineage: 'Ward One' was a 100 kWt TRISO HTGR planned for under a month of operation
2025-04-07 · orig-017 · Valar Atomics, Isaiah Taylor, Ward One

The April 2025 lawsuit post describes Ward One as a 100 kWt high-temperature gas reactor using TRISO fuel, named, the founder writes, after a Manhattan Project-era relative of his, with a planned operating life of less than a month. It says the design relies on strong negative thermal reactivity feedback and low power density for fully passive decay-heat removal, and claims that holding its spent fuel for five minutes would give the same radiation exposure as a CAT (CT) scan. The later Ward 250 carries the same 100 kWt figure (orig-011, orig-034), which suggests Ward 250 grew out of the Ward One concept; at its September 29, 2026 town hall Valar said Ward One was renamed Ward 250 (rnm-001). The post says Valar was building and testing Ward One in the Philippines. For scale: a CT scan is about 1 to 10 millisieverts (FDA), so five minutes would need 12 to 120 millisieverts an hour; the post gives no amount of fuel, cooling time, distance or shielding.

In plain termsWard One was Valar's first planned uranium reactor: very small (100 kilowatts of heat) and meant to run for less than a month. It was to be built in the Philippines; in September 2026 Valar said Ward One was renamed Ward 250, its test reactor in Emery County (rnm-001).
verified 2026-10-02: Second check, late September 2026: post re-read; 100 kWt, TRISO, under-a-month life, passive decay-heat wording and the spent-fuel claim match (the post says 'CAT scan'). Quote exact; relative still unnamed. Revised before publication (late September 2026): added the CT-scan dose-rate scale (12-120 mSv/h over five minutes). Revised before publication (2026-09-30): reworded the detail in neutral terms. Corrected in V1.1 (October 2026): the Ward One status rests on Valar's post; see Corrections.
company claimreported In April 2025 Valar said it would build and test Ward One in the Philippines because of NRC rules
2025-04-07 · orig-018 · Valar Atomics, Philippine Nuclear Research Institute

The same April 2025 post says Utah, Texas and other states had expressed interest in hosting the test reactor, but that because the NRC had not exempted small test reactors, Valar was building and testing it in the Philippines instead. Four months later (August 2025) DOE's pilot selection gave Valar a U.S. path, and its first reactor to be built, Ward 250, went to Utah instead (orig-006); Ward One stayed a Philippine plan. The November 2025 NOVA release still lists the Philippine partnership, and ANS reported that month that Ward One was slated to be built there by a Philippine-incorporated Valar subsidiary. Set beside the record: as of April 2026, Philippine reporting says the reactor had not been built and the project was still seeking approval to use a site on University of the Philippines land (orig-064).

In plain termsIn April 2025 Valar said U.S. rules pushed it to test its first reactor abroad, in the Philippines. Four months later DOE selected Valar for its faster pilot program, and a different reactor, Ward 250, was built first, in Utah. The Philippine project was still waiting on a site approval in April 2026 (orig-064).
verified 2026-10-15: Second check, late September 2026: post and NOVA release re-read; ANS's November 2025 report, cited in the detail but missing from sources, added. Wording fixed: Ward One did not move to Utah; a different reactor (Ward 250) was built there.
recordreported Ward One in the Philippines: made public in early 2025, not built by April 2026, still seeking approval to build on university land
2025-03 · orig-064 · Valar Atomics, Valar Atomics Research Institute, Philippine Nuclear Research Institute, Greenpeace Philippines

March 2025 (GMA News and Daily Tribune, March 24; Power Philippines): speaking as president of Valar Atomics Research Institute, Kip Mock said Valar would build a 100 kW micro high-temperature gas reactor with the Philippine Nuclear Research Institute (PNRI) as a proof of concept, not connected to the grid and not generating electricity, at a cost in the 'tens of millions' of dollars; licensing with PNRI's regulatory arm had begun; he would not give a timeline or location. A US firm, EoS Organization, gave a 192-million-peso grant to the Philippine Nuclear Science Foundation in support of the reactor, and Mindanao State University-Iligan Institute of Technology joined as a partner. TechCrunch (February 2025) reported the PNRI contract covered a test-scale reactor and two full-scale reactors. Greenpeace Philippines condemned the deal (March 27, 2025). ANS (November 2025) described Valar Atomics Research Institute as a Philippine-incorporated subsidiary. The Philippine Collegian, the University of the Philippines student paper (April 2026), reports PNRI Director Carlo Arcilla said the project had Department of Science and Technology approval, aimed to finish within two to three years, and was seeking approval from the University of the Philippines to build within PNRI, which sits on UP Diliman land, and that the scientists' group AGHAM opposed it. Set beside Valar's April 2025 statement that it was building and testing this reactor in the Philippines (orig-018): a year later no reactor had been built there. Valar's claim of the world's first binding Coordinated Research Project with PNRI (orig-066) could not be checked: PNRI's website could not be read.

In plain termsBefore Utah, Valar planned its first uranium reactor, Ward One, for the Philippines, saying U.S. rules were too slow. It made the project public in early 2025. A year later the reactor still had not been built; the Philippine nuclear institute's director said the aim was to finish in two to three years, on University of the Philippines land, and local scientists and students objected to a test reactor in an urban area.
verified 2026-10-17: Second check, late September 2026: GMA, Daily Tribune, Power Philippines (dated March 24, 2025), Greenpeace, ANS, Philippine Collegian, TechCrunch and the Valar post re-read: 100 kW, proof of concept, no grid, 'tens of millions', licensing begun, P192M EoS grant, MSU-IIT, DOST approval, 2-3 years, seeking UP site approval. Match. The student quoted by the Collegian is not named here.
recordreported Feb 2025: Valar leaves stealth with an $18M-19M seed led by Riot Ventures and a Philippine pilot-site contract
2025-02-20 · orig-020 · Valar Atomics, Riot Ventures, AlleyCorp, Initialized Capital, Day One Ventures, Steel Atlas, Philippine Nuclear Research Institute · seed round (reported): $19,000,000

TechCrunch (Mike Butcher, February 20, 2025) reported Valar had raised a $19 million seed round led by Riot Ventures, with AlleyCorp, Initialized Capital, Day One Ventures and Steel Atlas, to develop its first test reactor. It reported an initial contract with the Philippine Nuclear Research Institute to build a reactor in that country, and quoted the company saying its thermal test unit took about 10 months to design and build. The article described the technology as helium-cooled and reaching up to 900 C. It did not say where the company is based. Initialized Capital, an investor, gave the round as $18 million; the two figures are compared in fund-003.

In plain termsValar came out in public in February 2025 with $18-19 million from investors (published figures differ) and a deal to build a reactor in the Philippines.
verified 2026-10-02: Second check, late September 2026: TechCrunch re-read: $19M seed, investors, PNRI contract, ~10-month thermal unit, helium to 900 C, no base city named. All match.
company claimconfirmed Ward Zero, a full-scale non-nuclear thermal test reactor, completed 15 Feb 2025 (company date)
2025-02-15 · orig-005 · Valar Atomics, Ward Zero

Valar's Ward 250 page lists 'Ward Zero Completion' on 02.15.2025. The technology page describes Ward Zero as a 1:1 non-nuclear thermal test reactor whose core was fitted with silicon carbide heating elements so every system could be tested at full operating temperature. The company's August 2025 post places the WardZero prototype in Los Angeles and says it had then undergone five months of heat and pressure testing with two maintenance outages. Ward Zero was tested without nuclear fuel.

In plain termsWard Zero was a full-size practice reactor with electric heaters where the uranium would go, used to test the machinery at working temperature. It was tested without nuclear fuel; by Valar's own safety papers, Ward 250 reuses parts of it. The completion date is the company's own, and no Ward Zero test report was found in public records as of late September 2026 (orig-058, orig-059).
verified 2026-10-15: Second check, late September 2026: Ward 250 page date, technology-page description and the August 2025 post (Los Angeles, five months, two outages) re-read; all match.
recordconfirmed What became of the leaders' earlier reactor ventures: South Africa stopped funding PBMR in 2010; Ultra Safe Nuclear went bankrupt
2024-12-18 · ppl-015 · Pebble Bed Modular Reactor (Pty) Ltd, Ultra Safe Nuclear Corporation, NANO Nuclear Energy, Mark Mitchell, Valar Atomics · price NANO Nuclear agreed to pay for USNC reactor technology in bankruptcy: $8,500,000; approximate spending on South Africa PBMR, 2000s (ZAR 9.244 billion): $1,300,000,000

Valar's Chief Nuclear Officer, Mark Mitchell, came from two helium-cooled, coated-particle-fuel reactor ventures that did not build a power reactor; an early investor, Initialized Capital, wrote in February 2025 that Valar's team included 'several key USNC team members'. (1) Pebble Bed Modular Reactor (PBMR), South Africa, where, per Valar's release, Mark Mitchell's nuclear career began: World Nuclear News (20 Sept 2010) reported the government had decided 'no longer to invest' after about ZAR 9.2 billion (about $1.3 billion) over a decade, mostly public money, having found no anchor customer or investment partner and facing at least ZAR 30 billion more in costs and repeated missed deadlines; the project was to be placed under care and maintenance. The minister said the validity of the pebble-bed technology itself was not in doubt. (2) Ultra Safe Nuclear Corporation (USNC), where Mitchell established and led the Micro Modular Reactor program and, per Deseret News and the Sept 2025 OED/Valar release, served as president (ppl-003): Nano Nuclear Energy announced on 24 Dec 2024 that it would buy USNC's MMR and Pylon reactor technology and related partnerships for $8.5 million in cash, through a court-supervised auction in USNC's pending Chapter 11 bankruptcy in the U.S. Bankruptcy Court for the District of Delaware, approved on 18 Dec 2024; the release describes the MMR system as still 'under development'. Mitchell's move to Valar was announced in April 2024, about eight months before the court approved that sale. None of this is a finding against any individual; it is the track record of the programs.

In plain termsValar's Chief Nuclear Officer came from two earlier helium-cooled reactor projects. Neither got to a working power plant: South Africa's government stopped funding one after about $1.3 billion had been spent, while saying the technology itself was sound, and the other went bankrupt, and a bankruptcy court approved the sale of its reactor designs for $8.5 million.
verified 2026-10-17: Re-read World Nuclear News (20 Sept 2010), NANO Nuclear's release (24 Dec 2024), Valar's release announcing Mark Mitchell (29 Apr 2024), Deseret News (17 Mar 2026), the September 2025 groundbreaking release, Initialized Capital's post (19 Feb 2025) and a Sandia National Laboratories report for DOE (June 2023).
recordconfirmed Early website: a 2024 recruiting page, then 'thousands of reactors on one large site' making fuels, hydrogen and power
2024-12-06 · orig-062 · Valar Atomics

Internet Archive copies of Valar's site show how it first presented itself. April 2024: a one-page 'Engineers Wanted' recruiting site whose only updates are the company's unveiling and a $1.5m pre-seed from Riot Ventures. December 2024, 'Atomic Energy Reborn': Valar builds thousands of reactors on one large site and sells grid power plus hydrogen from a sulfur-iodine cycle, data center power, industrial power, and synthetic hydrocarbon fuel made by combining its hydrogen with CO2 captured from coal and natural gas plants and from direct air capture, through a modified Fischer-Tropsch process. The team listed then: Isaiah Taylor (Founder and CEO), Mark Mitchell (Chief Nuclear Officer), Willem Van Rooyen (Head of Mechanical Engineering), Visura Pathirana, PhD (Lead Nuclear Engineer), Kip Mock (Head of Operations) and Max Ukropina (Head of Project Development); address El Segundo. January 2025 mission page: 'hundreds of reactors on one Gigasite', three leaders with biographies (the mechanical head's cites leading heat-supply design for Ultra Safe Nuclear's MMR) and 18 staff, among them Max Ukropina, listed under 'Project Development'. September 2026: the homepage says hundreds, the leadership list has five names including a President (orig-003), and the mission page no longer lists staff. In the copy read on 17 October 2026 the leadership list had four names: Kip Mock, listed as Head of Operations in the December 2024, January 2025 and September 2026 copies, was no longer on the page. Valar's California filing gives its business as nuclear heat for fuel synthesis (orig-052).

In plain termsValar began as a small recruiting website in 2024. By late 2024 its pitch was to cluster thousands of reactors on one site and use their heat to make hydrogen and synthetic fuels, including from coal-plant carbon dioxide. By 2026 the pitch was 'hundreds' of reactors and leaned toward power for AI data centers (orig-063). As of September 2026, Valar's mission page listed five leaders and no staff; by 17 October 2026 it listed four, without Kip Mock.
verified 2026-10-17: Second check, late September 2026: April 2024, December 2024 and January 2025 archives and the live mission page re-read: recruiting page, 'thousands of reactors on one large site', S-I hydrogen, Fischer-Tropsch fuel, team list, El Segundo address, MMR bio, 18 staff, 'hundreds' now. Match; quotes exact.
recordconfirmed Valar's 'Ward Zero Whitepaper' page holds no Ward Zero data: its text is a 2019 essay by another author, word for word
2024-12-01 · orig-058 · Valar Atomics, Cairo Review of Global Affairs, Ward Zero

Valar's website has a page titled 'Ward Zero Whitepaper', subtitled as a detailed account of how its first prototype reactor works, dated December 1, 2024 and credited to Isaiah Taylor and two others. It is not listed in the library but loads at its public URL (read in late September 2026), and its entry is carried in the data that the Ward 250 page loads. Its body says nothing about Ward Zero. Every one of its body sentences matches, word for word, 'A Nuclear Curse and a Nuclear Blessing', an essay on Russian and Chinese reactor exports to the Middle East by Selim Sazak in the Fall 2019 issue of the Cairo Review of Global Affairs; the page names no source and still says the UAE's Barakah plant is expected to be fully operational by 2020. Archived copies show the same text on the page in April 2025, beside three other unlisted posts whose bodies read only 'Lorem ipsum', and again in October 2025, by when the placeholder posts were gone (orig-071). That fits leftover placeholder text rather than a finished paper (no Valar statement on it was found); the record does not show why it was used or why it remains. Either way, the one document Valar titled as a technical account of Ward Zero contains none (orig-059).

In plain termsValar's safety documents lean on tests of its non-nuclear prototype, Ward Zero. The only page on its website titled as a Ward Zero 'whitepaper' contains no Ward Zero information; its text matches, word for word, a 2019 magazine essay by another author about Middle East politics. The page is not listed in the site's menus and may be leftover placeholder text (no Valar statement on it was found), but it means no technical paper on Ward Zero was found in public records (as of late September 2026).
verified 2026-10-02: Second check, late September 2026: live page, Cairo Review essay (Selim Sazak, Fall 2019) and April and October 2025 archives re-read; all body sentences found verbatim in the essay. Cross-reference fixed (the placeholder posts are in orig-071) and the October 2025 capture, which the detail relied on, added. Revised before publication (late September 2026): wording on the unlisted page and the 2019 essay restated; placeholder reading attributed as unconfirmed; source title parenthetical now 'not listed in Valar's library'.
recordreported California registration (third-party copy): Delaware-formed Valar Atomics Inc. registered in California on March 5, 2024
2024-03-05 · orig-052 · Valar Atomics, California Secretary of State

California's bizfile system could not be read (bot protection), so this rests on bizprofile.net, a site that republishes state filings and says its data was extracted from the California Secretary of State's registry as of March 25, 2026. It shows Valar Atomics Inc., a stock corporation formed in Delaware, filed in California on March 5, 2024 under document number 6130344, status active and in good standing; principal address 4857 W 147th St, Hawthorne; a mailing address at a Wilmington, Delaware office suite; registered agent Firstbase Registered Agent Inc.; business type 'Nuclear Heat for Fuel Synthesis and other applications'. The officers listed are Taylor as chief executive, chief financial officer and secretary; the latest Statement of Information on the copy was filed November 19, 2024, which also moved the principal address from Wilmington to Hawthorne, and it predates Valar naming a separate President (orig-003, ppl-005). Registering in California about eight months after the stated July 4, 2023 founding is ordinary for a Delaware company that starts doing business in California; it neither confirms nor contradicts the founding date (orig-053).

In plain termsA company formed in one state has to register in each other state where it does business. This copy of Valar's California registration says it was formed in Delaware, registered in California in March 2024, and describes its business as 'nuclear heat for fuel synthesis', its original pitch. Because this is a private website's copy rather than the state's own page, it is marked as reported, not confirmed.
verified 2026-10-15: Second check, late September 2026: bizprofile copy re-read: Delaware-formed, filed March 5, 2024, doc 6130344, active, good standing, addresses, agent, business type and filing history. Officer-date wording made exact. Still a third-party copy: stays 'reported'.
company claimreported Company's own origin account places its founding about three years before August 2026
2023 · orig-015 · Valar Atomics

The August 2026 Series B post says Valar was 'born' three years earlier and that Ward 250's completion came less than three years after founding, implying a 2023 founding. It says the founding ethos was to bring manufacturing economics to nuclear through vertical integration and a focus on hardware. The post also says the NOVA core took two years to complete and 'It took seven months to take Ward 250 critical.' Read in sequence, seven months matches the time from NOVA's criticality (November 17, 2025, orig-008) to Ward 250's (June 18, 2026); the nine months Taylor gave in DOE's release counts from the empty site, i.e. the September 2025 groundbreaking (orig-007). The two figures count from different starting points and do not conflict, though the post does not say which start it means.

In plain termsValar says it was founded about three years before August 2026, which fits a mid-2023 start. It gives two lengths for the Utah build: 'seven months' to criticality in one post and 'nine months' elsewhere. Seven months matches the time since its test core first went critical in Nevada; nine months is the time since it broke ground in Utah. Both fit the company's own dates.
verified 2026-10-15: Second check, late September 2026: Series B text re-read. Earlier wording said the seven-month figure fit no dated milestone; it fits NOVA criticality (Nov 17, 2025) to Ward 250 criticality (Jun 18, 2026) exactly. Corrected.
analysisreported Public launch and $1.5M pre-seed from Riot Ventures date to late 2023
2023-11 · orig-050 · Valar Atomics, Riot Ventures, Isaiah Taylor · pre-seed round (reported): $1,500,000

Valar's press library lists two early items by Taylor: 'Announcing Valar Atomics' (an X post) and 'Announcing a Pre-Seed Funding Round Led By Riot Ventures' (a LinkedIn post). Neither post could be opened (both sites need a login). The library's own data dates them November 3, 2023 and December 6, 2023; the LinkedIn post's ID, decoded, points to November 8, 2023 (an inference). The earliest archived copy of the company's homepage found (April 2024) lists 'Unveiling Valar Atomics' and 'Announcing a $1.5m pre-seed investment from Riot Ventures' as its only updates (orig-062). Mother Jones reports the pre-seed was $1.5 million from Riot Ventures. Riot Ventures later led the $19M seed (orig-020).

In plain termsValar unveiled itself in November 2023 and, by the end of that year, announced $1.5 million of early money from Riot Ventures, which later led its $19 million round.
verified 2026-10-15: Second check, late September 2026: library data (Nov 3 and Dec 6, 2023), April 2024 archive ('Announcing a $1.5m pre-seed investment from Riot Ventures.') and Mother Jones ($1.5M from Riot) re-read. The LinkedIn ID decodes to Nov 8, 2023 and the X post ID to Nov 3, 2023 (both inferences). Match. Revised before publication (late September 2026): Mother Jones source title shortened to author and date.
recordreported Contrary Research dates the founding to July 2023 and places HQ in El Segundo, CA (Contrary is an investor)
2023-07 · orig-021 · Valar Atomics, Contrary Research, Isaiah Taylor

Contrary Research's company profile (updated November 2025) gives a founding date of July 2023, headquarters in El Segundo, California, about 45 employees and $150M total funding at that time, and says Valar emerged from stealth in early 2025. The same page discloses that Contrary is an investor in Valar, so it is not independent. Its July 2023 date is consistent with Valar's own 'three years ago' framing in August 2026 (orig-015). Delaware's registry, which would show the incorporation date, sits behind a CAPTCHA and was not read (orig-053).

In plain termsAn investor's research page says Valar was founded in July 2023 and is based in El Segundo, California. Because the publisher invested in Valar, it is not an independent source; the address records are in orig-036.
verified 2026-10-15: Second check, late September 2026: Contrary profile re-read: Jul 2023, El Segundo, $150M, 45 employees, stealth exit early 2025, updated Nov 11 2025, investor disclosure. All match; quote exact.
recordreported Deseret News: Taylor founded Valar on July 4, 2023; company is El Segundo-based with a Utah operations director
2023-07-04 · orig-029 · Valar Atomics, Isaiah Taylor, Jess Housekeeper, Deseret News

A Deseret News profile (Eva Terry, March 17, 2026) says Taylor founded Valar Atomics on the Fourth of July in 2023 and calls it an El Segundo-based startup; it names Jess Housekeeper as Valar's director of Utah operations and says the reactor's stackable concrete shielding blocks are made in Salt Lake City. This agrees with Contrary Research (an investor in Valar), which gives July 2023 and an El Segundo headquarters (orig-021); an unfinished June 2026 page on Valar's site gives a Hawthorne, California headquarters (orig-059). No state record showing the formation date could be read (orig-053).

In plain termsA Utah newspaper reports that Valar was founded on July 4, 2023. The article does not say where the date comes from, and no state record showing the formation date could be read (orig-053).
verified 2026-10-02: Second check, late September 2026: Deseret profile re-read: July 4, 2023, El Segundo, Jess Housekeeper as director of Utah operations, shielding blocks made in Salt Lake City. All match. Plain no longer asserts the date's origin. Corrected in V1.1 (October 2026): the Delaware line is removed; see Corrections.
analysisreported Founding date: July 4, 2023 is the company's account; the earliest records found are late-2023 posts and a March 2024 filing
2023-07-04 · orig-053 · Valar Atomics, Isaiah Taylor

Claim: Deseret News (March 2026) reports that Taylor founded Valar on the Fourth of July, 2023 (orig-029); Contrary Research, an investor, says July 2023 (orig-021); Valar told an Emery County town hall in July 2025 that it was about two years old (orig-045); its August 2026 Series B post says it was born three years earlier (orig-015). Record: the earliest dated items are the company's own library entries for Taylor's 'Announcing Valar Atomics' post (November 3, 2023) and pre-seed post (December 6, 2023, per the library; orig-050); the earliest archived copy of the website found (April 7, 2024) is a recruiting page listing only those two updates (orig-062); and the California registration of the Delaware corporation is dated March 5, 2024 (orig-052). The Delaware certificate of incorporation, which would show the legal formation date, could not be read (CAPTCHA). Nothing read contradicts a mid-2023 start, and nothing independent confirms July 4; the date doubles as the company's patriotic theme (Ward 250 is named for the nation's 250th birthday, orig-045). Taylor's pre-Valar career, as an investor describes it, included software work for the Department of Defense and Bridgewater Associates (Initialized Capital, February 2025); that is the investor's account; Taylor told Deseret News he briefly worked for a hedge fund and did engineering work for the Department of Defense (ppl-001).

In plain termsValar says it was founded on the Fourth of July, 2023. The earliest public traces found are the founder's own announcement in November 2023 and a California filing in March 2024. That fits the story but does not prove the exact day; the Delaware incorporation record would show the legal formation date, but as of September 2026 the state's online search required solving a CAPTCHA, so it was not read.
verified 2026-10-15: Second check, late September 2026: Deseret, library data, bizprofile and Initialized re-read; all match. Deseret's matching account of his earlier work added to the detail.
recordreported Isaiah Taylor, founder and CEO: by his and news accounts a self-taught coder and auto-shop co-founder; no nuclear-industry job found in records read (Sept 2026)
2023-07-04 · ppl-001 · Isaiah Taylor, Valar Atomics, Kip Mock, Erber Auto

Valar's mission page lists Isaiah Taylor as Founder & Chief Executive Officer (orig-003), and a House committee disclosure form for a 2 September 2026 hearing gives his position as 'CEO & Founder', representing Valar Atomics; its bio calls him a 'self-taught engineer and repeat founder' whose background spans cloud infrastructure, embedded systems and IoT platforms. His path before Valar, as told in the reporting read: Deseret News (17 Mar 2026) says he left high school at 16, co-founded an auto repair shop in Moscow, Idaho with Kip Mock (Erber Auto; both sold their shares when he started Valar, and it still operates), then started an automotive software startup that failed, briefly worked for a hedge fund, and in his words 'did some stuff for the Department of Defense on the engineering side'. Mother Jones (Feb 2026) cites Taylor's own 2024 post: self-taught coder, left high school, 'making six figures' by 16, auto shop at 17, a software career that 'did well'; the Utah Investigative Journalism Project (July 2025) also names the auto repair company founded with Mock. None of these name the software employer, the hedge fund or the Defense Department work, and no record of that work was found as of late September 2026. No account read shows a nuclear-industry job or nuclear degree before he founded Valar on 4 July 2023 (orig-029); Mother Jones describes him as having 'practically no experience with nuclear energy' (its assessment). The biography rests largely on Taylor's own telling.

In plain termsThe man in charge of Valar, who under the company's own quality-assurance program holds 'ultimate legal responsibility' as chief executive (safe-036), did not come from the nuclear industry on any record we found. By his own account he taught himself to program, left high school, and co-founded a car-repair business before starting Valar at 24. That is not against any rule. It means the company's technical claims rest on its engineers and on the Energy Department's review, not on the founder's own training.
verified 2026-10-17: Re-read Deseret (17 Mar 2026), Mother Jones (26 Feb 2026), UIJP (8 Jul 2025) and the mission page. Holds: dropout at 16, Moscow auto shop with Mock (Erber Auto), failed auto-software startup, hedge fund, DoD quote, founding 4 Jul 2023, 'at 24'. Headline now attributes the biography to his and news accounts; plain softened (safe-036 cross-checked). Personal matters in the sources outside the public role were left out. QAPD opened in late September 2026 to confirm the CEO's 'ultimate legal responsibility'; added as a source. Revised before publication (late September 2026): the exclusion note no longer names the personal matters it excludes; Mother Jones cited as 'Mother Jones (Kiera Butler), 26 Feb 2026' (narrow quote; headline not repeated).
company claimconfirmed Valar describes itself as scaling nuclear for heavy industry and clean fuels via multi-reactor 'gigasites'
undated · orig-001 · Valar Atomics

The company's own homepage (read in late September 2026) says Valar is scaling nuclear energy for heavy industrial power and clean hydrocarbon fuel production. Its stated model is to build hundreds of reactors on company 'gigasites' and sell grid-independent products: hydrogen, data center power, heavy industrial power and clean hydrocarbon fuels. This is the company's self-description, not an independent assessment. The page describes gigasites as a plan; no source read shows a gigasite built as of late September 2026.

In plain termsValar says it wants to build hundreds of small reactors together on very large company sites it calls 'gigasites', and sell what they make: hydrogen, power for data centers and factories, and fuels. No built gigasite was found in public records as of late September 2026; the company has one small test reactor, in Emery County.
verified 2026-10-15: Second check, late September 2026: homepage re-read; wording and quote match. 'No gigasite exists' reworded to what the sources show.
company claimconfirmed Mission page: 'one standardized design built thousands of times', vertically integrated design-build-operate
undated · orig-002 · Valar Atomics

Valar's mission page frames its mission as 'Abundant Energy for All Mankind' and contrasts it with the fewer than 700 commercial reactors built since 1942 across dozens of designs. It says the company will build one standardized design thousands of times, put hundreds of reactors on one gigasite, and vertically integrate design, construction and operation to amortize site costs. These are stated intentions, not demonstrated outcomes.

In plain termsThe idea is mass production: one reactor design repeated thousands of times, like a factory product, instead of every nuclear plant being a one-off. It is a plan, not something shown to work at any scale yet.
verified 2026-10-15: Second check, late September 2026: mission page re-read; the 700-reactor comparison, design-thousands-of-times and gigasite wording match; quote exact. Revised before publication (late September 2026): wording restated.
company claimconfirmed Company-listed leadership: Taylor (founder/CEO), Shahzad (President), Mitchell (CNO), Ukropina; Mock listed 29 Sept 2026, not 17 Oct
undated · orig-003 · Valar Atomics, Isaiah Taylor, Muhammad Shahzad, Mark Mitchell, Kip Mock, Max Ukropina

The mission page listed five leaders as of late September 2026: Isaiah Taylor, Founder & Chief Executive Officer; Muhammad 'Mo' Shahzad, President; Mark Mitchell, Chief Nuclear Officer; Kip Mock, Head of Operations; Max Ukropina, Head of Projects. The copy read on 17 October 2026 lists four: the same names and titles without Kip Mock, who appears nowhere on that page; the page gives no reason for the change. The page says the team has decades of experience researching, designing and constructing HTGRs; in the copy read on 17 October 2026 each listed leader also has a short company-written bio in the page text (in the 29 September copy the bios sat in the page's embedded data, behind 'Read bio'). See the People tab for roles and histories.

In plain termsThese are the five people Valar named as its leaders as of late September 2026. By 17 October 2026 its mission page named four; Kip Mock was no longer listed, and the page gives no reason. Who they are and what they did before is in the People tab; how the published team changed since 2024 is in orig-062.
verified 2026-10-17: Second check, late September 2026: mission page re-read in late September 2026: the same five names and titles; quote exact.
company claimconfirmed Stated reactor basis: HTGR design principles with TRISO fuel, graphite moderator, helium coolant
undated · orig-004 · Valar Atomics

Valar's homepage says its reactor uses high-temperature gas reactor (HTGR) design principles and lists TRISO fuel, a graphite moderator and helium coolant. The technology page names the architecture 'Numenor' and says HTGRs are optimized for high-grade process heat for industrial and chemical uses. The company claims an 'unmatched' safety profile and proliferation resistance for HTGR plus TRISO; that is a company claim, and neither page gives power ratings, temperatures or enrichment. HTGR and TRISO are decades-old concepts, which the company itself acknowledges ('Advanced nuclear designs have been around for decades').

In plain termsValar's reactor is a gas-cooled type: helium gas carries the heat away, graphite slows the neutrons, and the uranium sits inside tiny coated beads called TRISO (see the TRISO explainer). The type is decades old, as the company's own site says. 'Unmatched' safety is the company's word, not a regulator's finding.
verified 2026-10-15: Second check, late September 2026: homepage and technology page re-read; HTGR, TRISO, graphite, helium, 'unmatched' and 'Numenor' wording match; both quotes exact.
analysisconfirmed Utah's energy-lab page calls June 18 'full-power' criticality; DOE and ANS called it zero-power
undated · orig-012 · Utah San Rafael Energy Lab, Utah Office of Energy Development, Valar Atomics, U.S. Department of Energy

The Utah San Rafael Energy Lab's project page (a State of Utah site, read in late September 2026) says Ward 250 achieved 'full-power fueled criticality' on June 18, 2026, describes it as a Generation IV helium-cooled, graphite-moderated HTGR, and says the design is meant to scale to a 5 MW version. DOE's release and ANS both describe June 18 as a zero-power criticality demonstration; a DOE slide prepared for the NRC Commission's 21 July 2026 meeting lists 'June 18, 2026 – Criticality achieved (zero power)' and full power on 2 July 2026; and Valar's own Castle Country page says full power came 'days later'. The state page's wording therefore conflicts with the federal record; it also says Valar's design is engineered to 'remain impossible to melt under extreme conditions', which is a design claim, not a demonstrated result.

In plain termsUtah's own lab web page says the reactor reached 'full-power' criticality on June 18. The federal record says it was a zero-power test that day (a DOE slide shown to federal regulators lists full power on 2 July), and even Valar says full power came days later. The state page overstates what happened on June 18.
verified 2026-10-17: Second check, late September 2026: Utah lab page re-read in late September 2026: still says 'full-power fueled criticality' on June 18, 5 MW scale-up and 'impossible to melt'; DOE release says zero-power. Conflict stands.
recordconfirmed Name and naming: 'Valar' is from Tolkien; the 'Ward' reactors honor the founder's Manhattan Project ancestor
undated · orig-022 · Valar Atomics, Isaiah Taylor

Contrary Research, an investor, says the company name comes from J.R.R. Tolkien's mythology, and that the Ward Zero and Ward One names honor the founder's great-grandfather, who worked on the Manhattan Project. Valar's own April 2025 post says Ward One is named after that relative, whom the founder describes as having worked at Oak Ridge's X-10 reactor. TechCrunch described the ancestor as a grandfather and 'a nuclear physicist'; the company's own account says great-grandfather. The relative is not named here. The other Tolkien names (Numenor, Valarin, Windlord) and the 'Ward 250' birthday name are in orig-069.

In plain termsThe company's name comes from The Lord of the Rings author J.R.R. Tolkien: in his stories the Valar are the powers who shaped the world. The 'Ward' reactors are named, the company says, for the founder's relative who worked on the Manhattan Project. The other Tolkien names are explained in orig-069.
verified 2026-10-15: Second check, late September 2026: Valar post (great-grandfather, X-10), Contrary (Tolkien; great-grandfather; Ward naming) and TechCrunch ('grandfather ... a nuclear physicist') re-read; all match. Relative stays unnamed.
analysisconfirmed Stated operating temperatures vary by source: 'up to 900°C', 'above 750°C', 'in excess of 950º C'
undated · orig-023 · Valar Atomics

Published temperature claims for Valar's HTGR differ. TechCrunch (Feb 2025) reported up to 900 C; Valar's own August 2025 post claims operating temperatures above 750 C; Contrary Research (Nov 2025) says in excess of 950 C. None of these sources gives a measured Ward 250 outlet temperature, and Valar CEO Isaiah Taylor has described the Ward 250 test reactor as a 100 kWt reactor (ANS, June 2026). The figures are design targets or characterizations, not demonstrated operating data. The high temperature matters because the company's hydrogen plan relies on the sulfur-iodine cycle, whose hottest step needs roughly 800-1000 C per the Contrary report. An investor, Initialized Capital (February 2025), went further and wrote that Ward One 'operates at 900°C', although Ward One had not been built (orig-064).

In plain termsHow hot the reactor can run matters because very high heat is needed to make hydrogen cheaply. The figures Valar and others have published differ (750, 900, 950 degrees C) and none is a measured result. Valar's own safety paper lists 650 degrees C as the normal outlet temperature and 750 as a goal (orig-040).
verified 2026-10-16: Second check, late September 2026: TechCrunch (900 C), Valar's post (>750 C) and Contrary ('in excess of 950º C'; S-I step 800-1000 C) re-read. Added the investor's 'operates at 900°C' claim for an unbuilt reactor; plain now gives both NSDA temperatures. Re-checked 16 October 2026: ANS (22 June 2026) added as the source for Taylor's 100 kWt description; the headline quotes keep each source's own degree sign.
analysisconfirmed Ward 250's size is described inconsistently: '5 MW' in airlift coverage vs 100 kWt test power
undated · orig-031 · Valar Atomics, World Nuclear News, POWER magazine, American Nuclear Society, Deseret News

Coverage of the February 2026 airlift (World Nuclear News, ExecutiveGov, Deseret News) called Ward 250 a 5 MW or five-megawatt reactor able to power about 5,000 homes. POWER magazine (June 19, 2026) describes it as 100 kWt initial test power, scalable to 5 MWe, and ANS attributes the 100 kWt figure to Taylor; the Utah lab page says the design is meant to scale to 5 MW. Read together, 5 MW appears to be a design or scale-up rating, while the tested reactor is described by Valar at about 100 kW thermal for continuous running, roughly fifty times smaller by the stated figures (POWER gives the 5 MW as electric output and the 100 kW is heat, so the gap in heat terms is larger). Valar's draft safety agreement (NSDA Rev 02) sets 'Maximum continuous power: 100 kWth' and allows 250 kWth for short (1-hour) runs; DOE's categorical exclusion describes Ward 250 as 'a 250 kilowatt thermal (kWth)' reactor. As of late September 2026, no DOE document stating the authorized power limit was found in the records read (see open questions).

In plain termsNews stories called Ward 250 a '5-megawatt' reactor. Valar's papers put the reactor actually tested at about 100 kilowatts of heat for continuous running, some fifty times smaller; DOE's environmental review describes it as a 250-kilowatt (thermal) reactor. The '5 MW' figure belongs to a bigger future version.
verified 2026-10-15: Second check, late September 2026: POWER ('100 kWt initial test power and scalable to 5 MWe'), WNN, Deseret ('five-megawatt') and ANS re-read; all match. Revised before publication (2026-10-01): wording made more exact against the cited records.
recordconfirmed Addresses on record: El Segundo and Hawthorne overlapped in 2024-25; a Torrance address is on the Carbon claim notice
undated · orig-036 · Valar Atomics

Business addresses given for Valar Atomics Inc. in records read: (1) 1325 E El Segundo Blvd, Suite A, El Segundo, CA: on the company's own website in December 2024 (archived copy) and as the client address on its federal lobbying registration effective December 8, 2025. (2) 4857 W 147th St, Hawthorne, CA: the principal address on the California registration from a Statement of Information filed November 19, 2024, per a third-party copy (orig-052); DOE's September 30, 2025 fuel-line release lists Valar in Hawthorne; ANS called it Hawthorne-based in November 2025; it is the recipient and place-of-performance address in the federal award record (orig-055) and the address given for the mark's owner in the March 2026 trademark application (orig-051); Valar's June 2026 draft calls Hawthorne 'our headquarters'. (3) 20333 Normandie Ave, Torrance, CA: the locator address on the Notice of Location for the RW 453 lode claim in Carbon County, September 12, 2026 (mine-453). Press kept describing Valar as El Segundo-based: Contrary (November 2025), Mother Jones and Deseret News (early 2026), ANS (June 2026) and TechCrunch (July 2026). Valar's own jobs board listed 71 of its 92 openings in Torrance, California, when read in late September 2026, with two listings describing 'our Torrance, CA headquarters'; read October 16 and 17, 2026, it listed 81 of 101 openings in Torrance, one of them with that phrase. The board gives no street address. As of late September 2026, Valar's website gave no street address.

In plain termsValar has used at least three Southern California addresses: El Segundo on its early website and lobbying papers, Hawthorne as its registered and 'headquarters' address from late 2024, and a Torrance street address that, in the records read, appears only on the Utah mining-claim notice; Valar's jobs board lists most of its openings in Torrance (81 of 101 on 17 October 2026), and its listings have called Torrance its headquarters. Using more than one office is normal for a growing company.
verified 2026-10-17: Second check, late September 2026: combined with a duplicate address record. Archive, LDA, DOE, ANS, draft page, Deseret, Mother Jones and TechCrunch re-read; claim notice checked (Torrance address). Correction: the California copy shows Hawthorne as principal address from a November 19, 2024 filing, so 'Hawthorne from late 2025' was wrong; El Segundo and Hawthorne overlapped. Revised before publication (late September 2026): Mother Jones cited as 'Mother Jones (Kiera Butler), 26 Feb 2026' (narrow quote; headline not repeated).
recordconfirmed Legal entity: SEC filings name 'Valar Atomics, Inc.'; its QA and safety papers also say 'Valar Atomics Limited'
undated · orig-037 · Valar Atomics, Inc., TriplePoint Venture Growth BDC Corp., Cameco, CGF2021 LLC

A 10-K filed with the SEC by TriplePoint Venture Growth BDC Corp. (filed March 4, 2026) lists a growth capital loan to and common-stock warrants in 'Valar Atomics, Inc.' in its energy holdings as of December 31, 2025, and Cameco's Q3 2025 report names 'Valar Atomics Inc.' among DOE pilot selectees. Valar's own QA Program Description calls the company 'Valar Atomics, Inc.' in its text, while the covers of that document and of the NSDA say it is the property of 'Valar Atomics Limited', a name that appears in no registry record read. EDGAR shows nine investment vehicles named for Valar filing Form D in 2025-2026 (orig-056); no Form D filed by Valar itself was found in EDGAR as of late September 2026. State of incorporation is now confirmed from a federal record: the USPTO file lists Valar Atomics Inc. as a Delaware corporation (orig-051). A Philippine subsidiary, Valar Atomics Research Institute, is reported (orig-064). A search of the UK Companies House register (2026-09-29) returns no company named Valar Atomics Limited; where that name comes from is not shown in any record read.

In plain termsThe company's legal name is Valar Atomics Inc., a Delaware corporation (orig-051). Two of its own safety papers also say the documents belong to 'Valar Atomics Limited', a name found in no registry we could read.
verified 2026-10-15: Second check, late September 2026: TriplePoint 10-K (filed 2026-03-04; growth capital loan and common-stock warrants, as of 2025-12-31), Cameco Q3 2025 exhibit (filed 2025-11-05), QA PDF ('Valar Atomics, Inc.' in text; 'Valar Atomics Limited' on covers of QA and NSDA), EDGAR search and Companies House search re-read. All match.
analysisconfirmed What is borrowed: US AGR TRISO data, German AVR/THTR and HOBEG experience, NRC research-reactor rules
undated · orig-039 · Valar Atomics, AGR fuel program, AVR, THTR, HOBEG, Vallecitos Nuclear Center

Valar's own safety documents show how much of the design basis is inherited. The NSDA qualifies the fuel on US AGR-1/2 irradiation data and German AVR/THTR operating experience, adapts the safety method from NRC's research-reactor guide NUREG-1537 (used for university reactors), and cites the Vallecitos Nuclear Center as a precedent for a contractor-owned facility under DOE. The QA plan cites German HOBEG fuel-making technology. What the company presents as new is the business model (mass production, gigasites, vertical integration) and speed, not the reactor physics; the company itself notes advanced designs 'have been around for decades'. AVR and THTR burned pebble fuel, not the uranium-oxycarbide compacts Ward 250 uses, and the NSDA does not say which of their data apply.

In plain termsMuch of the reactor science Valar uses comes from earlier work: U.S. fuel tests, German reactors of past decades, and the rules for university research reactors. What Valar presents as new is the business approach and the speed.
verified 2026-10-15: Second check, late September 2026: NSDA (AGR-1/2, AVR/THTR, NUREG-1537 for 30+ university reactors, Vallecitos), QA plan (HOBEG) and technology page re-read. Match. Revised before publication (late September 2026): added that AVR/THTR used pebble fuel, not UCO compacts.
analysisconfirmed Fuel enrichment conflict: NOVA release says HALEU and 'same fuel' as Ward 250; NSDA lists 4.95% LEU
undated · orig-040 · Valar Atomics, Los Alamos National Laboratory

The November 2025 NOVA release says the NOVA core used HALEU TRISO fuel and 'the same fuel' as Ward 250. DOE defines HALEU as enriched between 5% and less than 20% U-235. Valar's October 2025 NSDA design table lists Ward 250 fuel enrichment as 4.95% U-235 under 'LEU limits', which is below the HALEU range. The October 2025 NSDA also gives two different outlet temperatures (650 C nominal in the design table, 750 C as a performance objective), and the company's NSDA web summary says testing up to 650 C. DOE's own shipping review (Safety Evaluation Report, approved by DOE's Headquarters Certifying Official on May 20, 2026) reviewed and recommended an amendment to allow limited shipments of HALEU TRISO fuel compacts, enriched to at most 19.9% U-235, from Los Alamos's criticality center in Nevada, with their end use as fuel elements in Ward 250 (safe-010). That matches the NOVA wording, not the NSDA's 4.95%. The NSDA says fuel design updates will be carried into the preliminary safety analysis (PDSA); no public copy of that analysis or of the final safety analysis was found as of October 17, 2026, so whether DOE's approved analysis covers HALEU cannot be checked (disc-001). A DOE slide prepared for the NRC Commission's July 21, 2026 meeting lists Valar's fuel as '19.75 wt% TRISO' and the two analyses' approval dates, but not their contents.

In plain termsValar's own papers disagree about the fuel grade: one says the richer HALEU grade, another says ordinary 4.95%. DOE's shipping review shows which fuel was cleared to be sent: HALEU (disc-001). A DOE slide shown to federal regulators in July 2026 lists the fuel as '19.75 wt% TRISO'.
verified 2026-10-17: Second check, late September 2026: NOVA release and NSDA re-read. The plain text relied on DOE's shipping review, which the fact did not cite: SER read and added, detail updated to match; DOE HALEU definition added.
analysisconfirmed Coverage arc: startup/tech press (2025), military airlift and politics (Feb 2026), critics (from March 2025), local (2026)
undated · orig-048 · Valar Atomics

Valar's own press library (read in late September 2026) curates the arc in 73 items (72 dated; one Chief Nuclear Officer announcement is undated), from the founder's 'Announcing Valar Atomics' post (November 3, 2023) to an interview dated February 28, 2026; nothing dated later had been added to that press list (the page's own announcements include an August 4, 2026 Series B post), although DOE, ANS, POWER, World Nuclear News and Deseret News covered the June 2026 criticality. The list runs: early-2025 launch coverage (TechCrunch, Axios, Business Insider) built on the founder's story and the $19M seed; March 2025 Philippine items; policy coverage of the May 2025 executive orders; August 2025 DOE pilot selection in trade and wire press; November 2025 NOVA criticality and the $130M Series A; a burst of national and military coverage of the February 2026 C-17 airlift; and local Castle Country outlets. It includes a few critical pieces (a Washington Post piece on political connections, an NPR safety story) but not the Utah Investigative Journalism Project, Mother Jones, Neutron Bytes, Greenpeace Philippines, The New Republic, ProPublica, the Philippine Collegian or the Partnership for Global Security critique. The critical reporting, in date order, is in orig-072.

In plain termsMost early coverage of Valar was launch, funding and airlift stories; criticism began in 2025 (a Greenpeace Philippines statement in March, a Utah investigation in July) and grew in 2026. Valar's own press list is mostly favorable and, as of late September 2026, listed nothing dated after February 2026 (orig-072).
verified 2026-10-15: Second check, late September 2026: library page data re-read: 73 press items, one undated; first 2023-11-03, last 2026-02-28; WaPo and NPR present; none of the named critical outlets present. Count and plain wording corrected.
analysisconfirmed From fuel synthesis to AI power: how Valar's emphasis shifted between 2024 and 2026
undated · orig-063 · Valar Atomics, NVIDIA

2024-2025: the California filing gives the business as nuclear heat for fuel synthesis (orig-052); the December 2024 homepage lists hydrogen, data center power, heavy industrial power and synthetic fuel from captured CO2 (orig-062); TechCrunch (February 2025) reported the aim of cheap hydrogen and synthetic fuels at up to 900 C; Valar's head of projects told the Trust Lands board that its 2024 Utah plan was to make fuels from coal-plant flue gas (mine-018, ppl-006); its December 2025 lobbying registration still describes 'clean hydrocarbon fuel production'. 2026: the July 1 event powered one NVIDIA chip and announced a 30 MW data-center study (orig-042); its BLM applications are titled 'Small Modular Reactor/Data Center' (mine-006); the Series B post leads with AI infrastructure; the Carbon County plan lists reactors, fuel manufacturing, hydrogen and synthetic fuels, data centers and steel (csite-001, csite-007). As of September 2026, no record read shows Valar producing any hydrogen or synthetic fuel, and its homepage still offers hydrogen through a sulfur-iodine cycle, a process that an investor-published profile says has never been deployed commercially (orig-024). Plans can change honestly. The record also shows continuity: data center power was on Valar's product list from December 2024, TechCrunch (February 2025) reported plans to power data centers and industrial plants, and the September 2026 homepage still says Valar is scaling nuclear energy 'for heavy industrial power and clean hydrocarbon fuel production'. The shift is in the headline product, from fuels toward AI power, while the fuel claims stay on the site.

In plain termsValar started out saying it would use reactor heat to make hydrogen and gasoline-like fuels. Data center power was on its list from 2024, and in 2026 its headlines are about powering AI data centers. The fuel and hydrogen plans are still on its website, but as of September 2026 no record read shows it has made either. That does not make the plan false, but it is still only a plan.
verified 2026-10-15: Second check, late September 2026: bizprofile, homepage, LDA registration and Series B post re-read; quotes exact. The TechCrunch report the detail relied on was missing from sources: added.
analysisconfirmed Valar's 'firsts', tested: one does not hold, some hold only narrowly, several cannot be checked
undated · orig-066 · Valar Atomics

(a) 'First criticality ever achieved by a venture-backed company' and, on its Castle Country page, the first startup in history to split the atom (Project NOVA, November 2025): the core ran on a Los Alamos machine operated by Los Alamos staff; holds only in that narrow sense (orig-047). (b) 'The first company ever to take a nuclear reactor critical outside of a national lab' (Series B, August 2026): does not hold as written; General Atomics, a company founded by General Dynamics in 1955 and not a national laboratory, commissioned its first TRIGA research reactor at its own site on May 3, 1958 (a 2008 conference abstract indexed by DOE's OSTI), and DOE's own wording is narrower, the first DOE-authorized reactor built outside a national lab, as is Valar's own Castle Country page, which says Ward 250 was the first reactor built and operated outside the national laboratory system under DOE's Reactor Pilot Program to reach criticality ahead of the July 4 deadline (ppl-030, orig-010). (c) 'The first-ever C-17 airlift of a nuclear reactor' (Castle Country page): holds as worded; broader 'first air transport' framing in coverage (a Reuters headline listed on Valar's own press page) does not, since the Partnership for Global Security notes earlier U.S. military air transport of small reactors, including two Russian TOPAZ-II space reactors flown by C-5 in 1992 (orig-073). (d) 'The first time in history a startup generated nuclear power' and 'the first and only time that an advanced reactor has directly powered AI infrastructure' (Series B): not testable; as of September 2026, no independent measurement of Ward 250's output was found in published records (orig-013). (e) The world's first binding Coordinated Research Project for an advanced reactor, with PNRI (state release boilerplate, September 2025): not checkable (orig-064). (f) Its reactor more photographed and filmed than any other in history (June 2026 draft): not testable. (g) Moving from concept to hardware faster than any team since the early days of nuclear (August 2025 post): not testable.

In plain termsValar often describes its milestones as historic 'firsts'. Checked against the record: the Los Alamos 'first' holds only narrowly, since a national lab ran the test; 'first company ever to take a nuclear reactor critical outside of a national lab' is not true, since a company (General Atomics) did it in 1958; 'first-ever C-17 airlift' is true as worded, though the military had flown reactors before on other planes; several others cannot be checked at all. This list shows which ones hold.
verified 2026-10-15: Second check, late September 2026: each cited page re-read; every quoted claim present and exact. The 1958 General Atomics counter-example had no source in this fact: OSTI record read and added; 'private company' reworded to what the record says.
analysisconfirmed The names: Valar, Numenor, Valarin and Windlord come from Tolkien; 'Ward' from the founder's family; '250' from 2026
undated · orig-069 · Valar Atomics

Valar: in J.R.R. Tolkien's writings the Valar are angelic 'Powers of the World' who entered the world to complete its making (Wikipedia, 'Valar'); Contrary Research says the company name comes from Tolkien (orig-022). Numenor: the name of Valar's reactor architecture on its technology page; in Tolkien, Numenor is an island kingdom west of Middle-earth, created by the Valar for Men in the Second Age and later destroyed in its Downfall. Valarin: the name of Valar's planned TRISO fuel plant in its QA program (orig-038); in Tolkien, Valarin is the language spoken by the Valar. Windlord: the February 2026 C-17 airlift, 'Operation Windlord' (orig-030); in Tolkien, Gwaihir the Windlord is a leader of the Great Eagles, who with other Eagles rescued Frodo and Sam from Mount Doom. Ward: the Ward reactors are named for a Manhattan Project-era relative of the founder; Valar says so of Ward One, and Contrary Research of Ward Zero and Ward One (orig-022). 250: Ward 250 marks the nation's 250th birthday, July 4, 2026, as Valar told Emery County (orig-045). 'Gigasite' is Valar's term for its reactor campuses, which TechCrunch compared to battery gigafactories. Valar has applied to trademark its company name (orig-051); whether it has filed for these product names could not be checked.

In plain termsValar names nearly everything after the world of The Lord of the Rings: the company (the godlike 'Powers' who shaped the world), its reactor design (Numenor, an island kingdom the Valar gave to Men), its fuel plant (the gods' own language) and its airlift (a great eagle). The reactor line itself is named for the founder's family and for America's 250th birthday. Names are branding, not evidence of anything, but they explain the words that turn up in the records.
verified 2026-10-15: Second check, late September 2026: Tolkien Gateway could not be read, so its four pages were removed as sources and Wikipedia's pages, read today, put in their place. Two meanings corrected to what the readable source says (Valarin is the Valar's own language; Gwaihir is 'a leader', not 'the greatest', of the Eagles). The trademark-search claim was removed (it could not be checked). Valar pages, QA PDF and TechCrunch re-read. Revised before publication (late September 2026): note wording simplified.
analysisconfirmed Critical reporting in order: the Philippines (2025), a Utah investigation (mid-2025), national critiques (late 2025-26)
undated · orig-072 · Valar Atomics

In date order, with what was read: March 27, 2025, Greenpeace Philippines condemns the PNRI deal, calling Valar a 2023 startup with no track record (read; orig-064); the scientists' group AGHAM criticized the project in a statement of March 24 (the Philippine Collegian, reporting it in April 2026, does not give the year; not read). April-May 2025, engineers dispute the 'CAT scan' claim online (as reported in the press; their own posts are not yet archived here; the claim: safe-026). July 8, 2025, the Utah Investigative Journalism Project with Utah News Dispatch, 'Who is Valar Atomics?': experts question its safety claims and note it had submitted no design to the NRC; Valar did not respond (read). July 23, 2025, E&E News on the NRC lawsuit (safe-028, csite-022). November 26, 2025, Washington Post, 'Kilowatts or connections?' (listed in Valar's own library; not read, subscription). December 17, 2025, NPR on safety worries (listed in Valar's library; not read, blocked). February 28, 2026, Neutron Bytes, 'Questions Abound About Valar Atomics' (read; orig-026). March 13, 2026, Partnership for Global Security, Ken Luongo (read; orig-073). March 20, 2026, ProPublica on DOE and the NRC. April 25, 2026, Philippine Collegian on opposition to Ward One (read; orig-064). May 26, 2026, The New Republic. July 9, 2026, MIT Technology Review on what criticality does and does not prove (read; orig-073). September 18, 2026, The Washington Sun on BLM data-center applications (mine-008). Locally, ETV News's first report of the Carbon County lease came the day after the state approved it (mine-041). Few public replies by Valar to these reports were found (ppl-029).

In plain termsThe earliest criticism of Valar found came from abroad, in the Philippines, in March 2025; it reached Utah in mid-2025 through an investigative nonprofit, and went national from late 2025 (The Washington Post and NPR, going by their headlines; neither was read) into 2026 (ProPublica, The New Republic and nuclear-policy critics). Few public replies from Valar to these reports were found. This list lets readers find each one.
verified 2026-10-15: Second check, late September 2026: Greenpeace, UIJP, Neutron Bytes, PGS, Collegian and MIT Technology Review re-read; dates and quotes exact; library dates for WaPo and NPR confirmed. The AGHAM statement's year is not given by the Collegian: wording fixed. Plain now counts the national 2025 pieces. Revised before publication (late September 2026): 'rarely answered' now 'few public replies were found'. Revised before publication (2026-10-01): wording made more exact against the cited records.
recordreported Funding lineage: $1.5M pre-seed, $19M seed, $130M Series A, a $2B-valuation raise, $1B Series B (reported $6B value)
undated · orig-057 · Valar Atomics, Riot Ventures, Sequoia Capital, Snowpoint Ventures · pre-seed, 2023: $1,500,000; seed, Feb 2025: $19,000,000; Series A, Nov 2025: $130,000,000; Mar 2026 raise ($340M equity, $110M debt), as reported: $450,000,000; Series B equity, Aug 2026: $1,000,000,000; credit facility, Aug 2026: $200,000,000

Rounds as announced or reported, in order: (1) $1.5 million pre-seed from Riot Ventures, announced late 2023 (company homepage, April 2024; orig-050). (2) $19 million seed led by Riot Ventures with AlleyCorp, Initialized Capital, Day One Ventures and Steel Atlas, February 2025 (TechCrunch; Initialized's own post says $18 million; orig-020). (3) $130 million Series A, November 2025, led by Snowpoint, Day One and Dream per ANS (orig-049). (4) March 2026: $450 million, of which $340 million equity and $110 million debt, at a $2 billion valuation, per a Bloomberg report cited by TechCrunch (Bloomberg not read). (5) August 2026: $1 billion Series B led by Sequoia plus a $200 million credit facility led by Erebor Bank and J.P. Morgan (company post; orig-014). TechCrunch says the company did not disclose a valuation and that Bloomberg put it at $6 billion; its July report says part of the new capital had been raised earlier at the lower valuation, so the amounts should not simply be added. Also on record: a growth capital loan and warrants from TriplePoint Venture Growth as of December 31, 2025 (orig-037) and about $8.1 million sold through nine investment pools (orig-056). No Valar financial statements were found in public records as of late September 2026; the only federal award found is $0 (orig-055); Utah approved up to $106.7 million in post-performance tax credits in July 2026 (csite-024).

In plain termsIn under three years Valar went from $1.5 million of seed money to more than $1 billion, at a reported value of $6 billion. Almost all of it is private money. Private companies do not publish accounts, so the public sees only what the company and reporters choose to say, plus a few government filings.
verified 2026-10-15: Second check, late September 2026: each cited source re-read: archive, TechCrunch ($19M), Initialized ($18M), ANS ($130M), TechCrunch July ($450M = $340M + $110M at $2B per Bloomberg; part of the new round raised earlier at the lower valuation), Series B post, TechCrunch August ($6B per Bloomberg). Headline marks the $6B as reported.