VALAR WATCHTake action

Their claims, annotated

Valar Atomics says
“The New Atomic Age”
Annotated, line by line, from their own documents and the public record.
11conflicts or unsupported claims
33unresolved
11claims that hold up
409sourced facts
646sources read
documented conflict
Valar's NOVA release says HALEU; its safety agreement says 4.95%.

The only Valar safety paper for Ward 250 found in public records lists 4.95% (ordinary) uranium. Valar's own Los Alamos release a month later calls the same fuel HALEU, and DOE's shipping review cleared shipments of HALEU of up to 19.9%…

See both sides and the sources →
documented conflict
Official: 'no dose at the site boundary.' Valar's worst case: under 100 millirem.

The state lab's director told local readers that companies must show no dose at the site boundary in an accident and that the public would 'never see a dose of radiation above background'.

See both sides and the sources →
documented conflict
Safety via the NRC, said the governor. NRC role: none on record.

Valar's own papers say DOE alone, and its staff's 'license' and 'permits' refer to the DOE authorization, as its safety paper's 'License limit' does. The statements that do not match the record are the state's: the governor, as KUTV…

See both sides and the sources →
They say
“Proven Technology”
Their fuel and design documents, read side by side.
documented conflict
Valar's NOVA release says HALEU; its safety agreement says 4.95%.

The only Valar safety paper for Ward 250 found in public records lists 4.95% (ordinary) uranium. Valar's own Los Alamos release a month later calls the same fuel HALEU, and DOE's shipping review cleared shipments of HALEU of up to 19.9%…

See both sides and the sources →
documented conflict
State lab: 'full-power fueled criticality,' June 18. DOE and Valar: zero-power.

The state lab's page says Ward 250 achieved 'full-power' criticality on 18 June, which DOE's release and Valar's own page contradict: 18 June was a zero-power test.

See both sides and the sources →
claim not supported
Hydrogen 'cheap and abundant.' Hydrogen produced to date: none on record.

The home page states, in the present tense, that Valar's reactors paired with its sulfur-iodine process produce cheap, abundant hydrogen. It can be read as a description of design intent ('When paired with'), but no hydrogen production is…

See both sides and the sources →
unresolved
76 fuel elements on paper. 24 cleared to ship. 'Very lightly loaded.'

The public paper's design core is 76 elements of 4.95% fuel. DOE cleared 24 elements of fuel up to 19.9% for shipment, and a Valar speaker calls the core 'very lightly loaded'; those two agree. What is actually in the core was not found in public records.

See both sides and the sources →
unresolved
250 kW: one hour per run in the safety paper; up to one day online.

The safety paper allows 250 kW for one hour per run; an unfinished, unlisted Valar page from June 2026 says up to one day. Read as a total across runs, the two can fit; read as one continuous day, it is 24 times the paper's limit.

See both sides and the sources →
unresolved
'Raw and uncleaned' data, said a draft page. Shared: one chart without power units or times.

The promise sits on a page that is unfinished and unlisted in Valar's library, so it may be a draft never meant to be read, which is a plausible simplest reading.

See both sides and the sources →
unresolved
'Directly powered AI infrastructure': one NVIDIA chip, at 37% power.

Valar's 'first and only' claims about powering AI and a startup generating nuclear power cannot be tested: no measured output was found in public records as of late September 2026.

See both sides and the sources →
They say
“Intrinsic Safety”
The safety paperwork, the numbers in it, and who checks them.
documented conflict
Official: 'no dose at the site boundary.' Valar's worst case: under 100 millirem.

The state lab's director told local readers that companies must show no dose at the site boundary in an accident and that the public would 'never see a dose of radiation above background'.

See both sides and the sources →
claim not supported
Five minutes holding spent fuel = a CAT scan, per Valar. No correction found on Valar's site.

Valar's five-minute CT-scan claim concerned Ward One, a planned 100 kWt reactor run for under a month, and gave no cooling time. Our check: it can hold for part of the fuel, or for all of it after about 10 to 130 days of cooling.

See both sides and the sources →
unresolved
Worst case assumes 3 megawatt-days. The limit line reads 30.

The accident inventory assumes 30 full-power days (3 MWd), as every other Valar statement of the plan does; one line of the same draft sets a 30 MWd limit. The likeliest simple reading is a unit slip; even ten times longer, iodine-131, the main dose source, would rise about 8%.

See both sides and the sources →
unresolved
Limit: 650°C. Goal: 750°C 'under normal operations.' Same document.

In the draft, 650°C is the outlet-temperature limit, and Valar's public summary says 650°C; 750°C appears as a goal. The likeliest reading is an out-of-date goal line; the approved operating limits were not found in public records.

See both sides and the sources →
unresolved
Public assumed 400 meters away, on paper. The lab: 20.6 acres.

The worst-case analysis in Valar's draft paper assumes no member of the public is closer than 400 m, which it calls 'the nearest point of public access'; no DOE document read sets 400 m as a required boundary, and DOE's approved analysis…

See both sides and the sources →
unresolved
Emergency plan: on-site only, requested. Off-site plan: none found.

Valar asked DOE to require only on-site emergency response and told DOE that state-DOE notification agreements and local drills would back it up. No DOE decision on the request, state-DOE agreement, emergency plan or drill record was found…

See both sides and the sources →
unresolved
Hands-off cooling target: 72 hours. DOE's review: 'indefinitely.' Analysis: not found in public records.

Valar's safety paper sets a minimum target of 72 hours with no operator action; DOE's categorical-exclusion review, under which no full environmental study was required, states that the passive system keeps fuel within limits indefinitely.

See both sides and the sources →
unresolved
'Regular environmental reporting' promised. Reports found in public: none.

Valar promised monitoring data and regular reports to state environmental agencies and an EPA air-emission evaluation. As of late September 2026, about three months into operation, none of it was found in public records; the only other…

See both sides and the sources →
unresolved
The 'Ward Zero Whitepaper': a 2019 essay about Middle East reactors.

Valar's safety paper credits Ward Zero tests for natural circulation and cavity cooling, and for filing its final safety analysis early. Valar has described the testing publicly only in summary.

See both sides and the sources →
unresolved
Future 'largest manufacturer' of TRISO. Today: pilot line, authorization not found.

Valar tells local councils and investors it will make its own fuel, at the largest scale in the country. Those are future goals that cannot be tested yet; DOE's own selection of Valar for its fuel-line pilot supports that Valar is building…

See both sides and the sources →
unresolved
Spent fuel destination: 'a DOE-approved facility.' Which one: unnamed.

Where Ward 250's spent core will go, when, and who pays if Valar cannot, are not stated in any record read. No state permit or license naming Valar was found as of late September 2026; both are conditions of Utah's 2026 exemption for…

See both sides and the sources →
unresolved
'Diligent in guaranteeing public safety.' Lab safety manager: posted Sept 25.

State officials describe Valar as diligent on safety, and a 2025 news article about the state lab says it is under strict supervision. Under the safety agreement the state can observe but not direct DOE's oversight, and DOE puts reactor…

See both sides and the sources →
unresolved
Worker-safety rule swap 'approved through the OTA.' The OTA: not found in public records.

Valar says the swap of DOE's worker-safety rule for OSHA standards during construction was approved through its DOE agreement, which was not found in public records; its own paper says the formal exemption was still in process, pending a…

See both sides and the sources →
unresolved
'Proliferation resistance when paired with TRISO.' Inside the TRISO: HALEU.

'Unmatched' is a comparison no record tests. On proliferation, the record points both ways: the fuel is HALEU, which a National Academies committee found potentially more attractive for weapons than ordinary reactor fuel, while noting the…

See both sides and the sources →
They say
“Four executive orders gave Valar a mandate to test our reactor.”
Who authorized it, under which rules, and what the public got to see.
documented conflict
Safety via the NRC, said the governor. NRC role: none on record.

Valar's own papers say DOE alone, and its staff's 'license' and 'permits' refer to the DOE authorization, as its safety paper's 'License limit' does. The statements that do not match the record are the state's: the governor, as KUTV…

See both sides and the sources →
unresolved
In no announced HALEU round. HALEU cleared to ship from Nevada.

Valar is in none of DOE's three announced HALEU allocation rounds, yet HALEU fuel for Ward 250 was cleared to ship from a federal site, most likely the NOVA fuel Valar says it provided.

See both sides and the sources →
unresolved
'DOE has approved,' said Valar. The PDF it linked: 'In Review.'

DOE confirms Ward 250 is DOE-authorized and that federal teams reviewed its readiness. The paper Valar said DOE approved was marked 'In Review'; a status line not updated is a plausible simple explanation. A DOE slide lists the two safety-analysis approval dates Valar gave; the other approvals' dates rest on Valar's word.

See both sides and the sources →
unresolved
'Same standards' as the NRC. Six requested substitutions. Approvals not found in public records.

Valar and DOE say the DOE route applies safety standards equivalent to the NRC's. The record shows a rewritten DOE rulebook; six substitutions Valar requested, two of which, both research-reactor standards (ANS-15.8 and ANS-15.4), Valar…

See both sides and the sources →
unresolved
2025: why engage the NRC? 2026: hiring a Director of Licensing.

In 2025 Valar was suing the NRC (a suit still open in September 2026, per a court status report the American Nuclear Society described) and saw no reason to engage it; Ward 250's safety paper rules out NRC licensing for that reactor.

See both sides and the sources →
They say
“Neighbors first.”
“Our doors remain open through tours, town halls and straight answers”
The Carbon County site: what is planned, what was filed, and what neighbors were told.
documented conflict
'Open with all plans' (Emery, April 7). Valar named with the Carbon site: August 10.

Valar promised openness about all its plans and describes an open-door policy. The public record shows the Carbon site mapped two days after that promise, noticed in April and July without Valar's name or any nuclear use, first tied to…

See both sides and the sources →
documented conflict
'Not trying to keep anything from anyone,' said the state's lab director. The MOU: confidential, except as required by law.

At Emery County's June 2025 hearing, the state's lab director said nothing was being kept from anyone; the state's agreement with Valar, a month old, was confidential under its terms. No record read shows what, if anything, was withheld.

See both sides and the sources →
documented conflict
Promised state-permit compliance. District minutes: tied in 'without prior authorization.'

Valar's safety agreement commits to complying with state permits; the local utility's approved minutes record a staff report that Valar connected to the county sewer before authorization and had not sent its sewer plans to the state as…

See both sides and the sources →
unresolved
DOE reviewed a 12-month test; a 10,270-acre application was pending.

DOE's review describes one short test with no expansion and states the proposal was not split up to fit an exclusion; its own text says phases with their own decisions are not segmentation.

See both sides and the sources →
unresolved
'Not hiding anything,' a commissioner said of Valar. No county agenda found named the site.

Officials described Valar as open and hiding nothing. Their words can fairly be read as praise of Valar's public education and its openness with the county and the board, which the record supports (the county worked with Valar from early…

See both sides and the sources →
unresolved
Notice names Valar; nearby mineral records list only gravel and CO2.

By law, a lode location is to follow the discovery of a vein. The area's public mineral records show only gravel, which cannot be claimed, and a gas obtained by lease; in Valar's favor, a claimant can know of a showing they do not list.

See both sides and the sources →
unresolved
'No burden on our water supply.' Also seeking up to 1,000 shares.

The reactor needs no cooling water, so that part of the claim holds, and Valar's 'waterless' wording refers to cooling. The site and campus plans do use water: Valar told Castle Dale about 10 acre-feet 'for industrial use', with no time…

See both sides and the sources →
unresolved
'Has also purchased,' says the state. 'Currently purchasing,' said Valar.

A state agency says Valar has bought the land; Valar itself said in mid-August that it was still waiting on clean title. ETV News had already called the land 'recently purchased' on 28 July, before Valar's August statements, so 'purchased'…

See both sides and the sources →
They say
“We designed and built our advanced test reactor in record time.”
The dates, as claimed and as recorded.
documented conflict
'Just a week later': thirteen days, by Valar's own dates.

Valar's Series B post says electricity came 'just a week' after criticality; its own website dates the two 13 days apart (18 June and 1 July). The two other mismatches first recorded here have simplest explanations: the '14 days' page is…

See both sides and the sources →
unresolved
A 12-month test on paper. '165 years' of fuel, per the CEO.

The safety, quality and environmental documents all treat Ward 250 as a one-time mission of about 12 months of operation followed by teardown, and use that short life to justify lighter requirements.

See both sides and the sources →
unresolved
'Building and testing this reactor in the Philippines.' A year on: unbuilt.

In April 2025 Valar said it was 'building and testing' its first reactor in the Philippines. The record shows a real, continuing project: a partnership with the Philippine Nuclear Research Institute, licensing begun (per Valar), a…

See both sides and the sources →
Analysis
What local audiences heard about scale: 'just testing' (July 2025) to 'hundreds of reactors' (Sept 2026)

In 2025 Emery County was told this was a small, one-year test, not a power plant, though Valar also spoke of commercial development by 2028. But that same year the local paper printed Valar's description of itself as building clusters of…

The evidence →
They say
“Castle Country was built on energy.”
Carbon and Emery County: meetings, votes, land and mining claims.
From the record
BLM's public mining-claim map shows claim names, serial numbers and status, but not who staked them or when

The federal claim map tells you that a claim exists, its name and number, its type and roughly where it is (to about a 160-acre square). It does not tell you who owns it.

The evidence →
From the record
43 CFR 3715: living on or fencing a mining claim needs BLM review first, and non-mining uses are prohibited

No claimant may live on a claim or keep structures on it for more than 14 days in 90, fence it off or keep people out without BLM's concurrence, or use it for non-mining purposes.

The evidence →
From the record
43 CFR 3809: small exploration needs only a notice; mining and bigger exploration need a plan with 30 days of public comment

Small-scale exploration on a claim can start after a notice to BLM, with no public-comment step in the rule read. Mining, and larger exploration, needs an approved plan, and those are published for comment.

The evidence →
From the record
BLM right-of-way rules: public meetings are held if there is enough public interest; notice comes with review

The rule does not by itself require public notice when a company applies to use public land. Public involvement comes mainly through BLM's environmental review and any public meetings BLM decides to hold, announced in the Federal Register.

The evidence →
They say
“Build with us”
Who pays whom, as filed: rounds and agreements.
unresolved
State cost 'virtually nothing.' State tax credit: up to $106.7 million.

In May 2025 the governor described the state's cost as virtually nothing beyond space and infrastructure, describing the MOU as it then stood. The record shows that the space is a publicly funded lab whose use terms were not found in…

See both sides and the sources →
unresolved
$1 billion raised, per Valar. Valar's own SEC filings found: none.

Valar's announced totals were not found in any Valar filing (as of late September 2026); the seed round is $19 million or $18 million depending on the source; valuations come only from press reports (Bloomberg as relayed, and TechCrunch's…

See both sides and the sources →
unresolved
County: out of the tax zone. Port authority: 'located within.'

The county voted to remove Valar's parcel from the port authority's tax-capture zone; the port authority's plan, adopted two days later, and its September statement still count it in.

See both sides and the sources →
unresolved
Parcel income: $4,000 a year, or a month. Same agency.

The memo says the parcel earns about $4,000 a year; the approved minutes say $4,000 a month. The meeting's transcript shows staff naming no period, so the minutes' 'per month' may be the minute-taker's addition and the memo's figure may be…

See both sides and the sources →
They say
“Join our mission”
Who is behind it, by public role and public record.
documented conflict
'First company ever' outside a national lab. General Atomics, 1958.

The Series B 'first company ever' claim is contradicted by General Atomics' 1958 TRIGA reactor at its own site; the narrower claims DOE and Valar make elsewhere (first DOE-authorized reactor built outside a national lab; first…

See both sides and the sources →
unresolved
'Relocated to the region from California.' 81 of 101 openings: California.

Valar told Carbon County it relocated from California, and told the state of about 40 full-time local residents and 275 future jobs. The record shows most of its openings and its registered place of work in California, and, in April, Valar…

See both sides and the sources →
unresolved
Documents: 'property of Valar Atomics Limited.' Federal records: Valar Atomics Inc.

The covers of the two public safety documents say they are the property of 'Valar Atomics Limited', a name that appears in no registry or federal record read; the DOE authorization and award name Valar Atomics Inc.

See both sides and the sources →
Analysis
Who holds the money: three investor directors on record, defense-tech backers, and a lender whose director is a backer

This is who has put money into Valar or lent it money. Several of the best-known names come from the defense-technology world: the founders or top executives of Anduril and Palantir and a Lockheed Martin board member; the largest round was…

The evidence →
Ask for the records
What the records would show, if anyone asked
Records requests, ready to send, are on the Take action page.
Fair is fair
Credit where due
Claims the record supports.
claim holds up
'Not a commercial power plant': correct. Next, as Valar told the state board: more commercial units.

None for Valar's claim: Ward 250 is a research reactor under DOE, as Valar consistently says. The wording that does not hold is others': a Carbon commissioner called it a 'commercial reactor', and the state lab's page speaks of moving…

See both sides and the sources →
claim holds up
Dropped, Valar told a reporter. BLM's record, Oct 6: closed.

Valar describes its Green River application as a survey-access request it has dropped. BLM's public record showed the case pending on 29 September, titled for a reactor and data center; that title and a survey-access request are consistent…

See both sides and the sources →
claim holds up
Waste 'strictly prohibited on the lease.' Unless Valar buys that part.

None as worded: waste storage is barred on land under the lease. The limit: the written memo in the same public board packet lets Valar buy any portion it wants for interim spent-fuel storage before storing anything there, which the…

See both sides and the sources →
claim holds up
Criticality before July 4, 2026: it happened. Credit where due.

None. The promised criticality came on 18 June 2026, ahead of the 4 July goal and of Valar's own 30 June target. DOE's release describes a zero-power criticality.

See both sides and the sources →
claim holds up
Can't melt: DOE agrees. Can't leak: not what the papers say.

None: 'cannot melt' matches DOE's own description of TRISO fuel. The limit of the claim is that 'cannot melt' does not mean 'cannot release radioactivity'; Valar's worst-case analysis assumes some particles fail and release gases, and a…

See both sides and the sources →
claim holds up
$0 in federal cash, as DOE said. Help in kind: unpriced.

None on cash: the record shows $0 in federal money, as DOE said. No public value for DOE's free review, the federal fuel route or the Los Alamos test was found in public records as of late September 2026.

See both sides and the sources →
claim holds up
First C-17 airlift of a reactor: holds as worded. Unfueled, too.

None as worded: no earlier C-17 reactor airlift was found in the records read as of late September 2026. Broader 'first air transport of a reactor' framing in coverage does not hold, and the flight carried no fuel.

See both sides and the sources →
claim holds up
'Seven months to take Ward 250 critical': checks out, from NOVA.

None as worded: both time spans match the record. 'Record time' was not shown in any record read as of October 2026; the fast spans are the Utah site and the step from the Nevada test, not the reactor hardware, which came from earlier work…

See both sides and the sources →
claim holds up
'Tours, town halls and straight answers': in Emery, often.

None for Emery County: Valar went before local bodies often and opened its site. This finding does not extend to the Carbon County site (csite-026), and minutes are summaries, so how complete the answers were cannot be judged.

See both sides and the sources →
claim holds up
No young faults under the Carbon parcel. The claim holds.

None: the applicant's statement matches the state's fault map. Its 'low-seismicity zone' description was not tested.

See both sides and the sources →
claim holds up
No Carbon County coal for years, said a commissioner. Correct.

None: the state's production figures support the commissioner's statement; they show no Carbon County output in 2020 through 2024, the latest year reported (mine-039), so if anything the period is longer than four years.

See both sides and the sources →
Analysis
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

Not everything Valar says is contradicted by the record. DOE did authorize the reactor, it did start up, and DOE says TRISO fuel is 'more resistant to neutron irradiation, corrosion, oxidation and high temperatures' than traditional reactor fuels.

The evidence →