Centrus Hit Its 2026 Build Milestones Early and Halved Anyway: The Risk Moved to a Lease Expiring in Eleven Months

Results vs. Consensus

Centrus reported after the close on November 4 and held its call the following morning, November 5, at 8:30 a.m. ET. This is our fifth consecutive note on the company and the fifth in which the share price and the underlying asset have moved in opposite directions. We initiated at Hold in November 2025 at a $277.78 reaction close, upgraded to Outperform in February at $210.16, maintained in May at $231.32 and again in August at $178.03, and write this at $89.03. The stock is down exactly 50.0% since our last note and 72.7% from the $325.73 close that preceded initiation.

The quarter itself is the least informative of the five and the filings around it are the most informative. Revenue, adjusted earnings and guidance all landed where they were expected to. What changed sits in a 10-Q published the morning of the call and in a warrant table that no earnings document mentions.

Q3 2026 Scorecard

The consensus band for a single quarter's revenue spanned 18% from low to high. That dispersion is not sell-side carelessness. It is the direct consequence of a business whose customers hold annual rather than quarterly purchase commitments, in a company that declines to guide quarterly. Q2 printed $176.1M and Q3 printed $55.5M off the same order book.

Year-over-Year Comparison

Sequential Comparison

Nine-Month Comparison

Quality of the Print

• Revenue: the composition is better than the headline and the level is worse. Separative work unit revenue grew 31.8%, its first increase of 2026, on a 19% volume increase that is also the first of 2026. Against that, $55.5M is the lowest quarterly revenue of the nine quarters back to Q3 2024, and the decline is the uranium line reverting from a $34.1M prior-year sale to $17.7M. There is no acquisition, no divestiture and no meaningful currency effect in the comparison.
• Margins: the loss is structural within the quarter and the quarter is structurally the weakest. Third-quarter operating results were $(7.6)M in 2024, $(16.6)M in 2025 and $(29.7)M now, and the current figure is the largest quarterly operating loss of the nine quarters back to Q3 2024. Roughly half of this year's $13.1M deterioration is the gross line and roughly half is operating expense: advanced technology costs rose $3.0M, with Growth Costs at $2.4M against $1.1M, and selling, general and administrative rose $3.7M, including $4.6M of warrant issuance costs. Excluding growth costs and stock compensation entirely, the adjusted operating loss still widened from $(14.9)M to $(24.0)M.
• EPS: there are three numbers and only two of them are about the business. GAAP income of $31.6M exists because the share price fell: the common warrants issued in September are liability-classified and remeasured each period, so their value dropped from roughly $266.8M at issuance to $220.4M at September 30 as the stock went from $181.49 to $139.49, and the $46.4M difference ran through the income statement. Normalized loss per share of $(0.49) removes that and the $4.6M of warrant issuance costs expensed into selling, general and administrative. Adjusted loss per share of $(0.46) further removes growth costs and stock compensation. The sell-side estimated the adjusted basis and hit it.

Segment Performance

Key Operating and Commercial Indicators

Low-Enriched Uranium segment

Segment revenue of $31.8M fell 29%, and the composition inside it moved in the direction the thesis has been waiting on for a year while the economics moved against it. Separative work unit revenue grew to $14.1M from $10.7M, on a 19% volume increase and a 10% increase in the average realized price. That is the first quarter of 2026 in which both volume and realized price rose together. The offset is that the average unit cost of each unit sold rose 28%, so the spread between the change in what Centrus charges and the change in what it pays was adverse for a second consecutive quarter and widened: plus seven points in Q1, minus ten in Q2, minus eighteen now. Uranium revenue of $17.7M against $34.1M accounts for the whole segment decline.

Management's framing is unchanged and remains correct as far as it goes.

"LEU customers generally have multi-year contracts that carry annual purchase commitments, not quarterly commitments. The gross loss in our LEU business varies based upon the timing of those contracts. The pricing applied to deliveries varies depending upon the market conditions at the time the contract was signed." — Q3 2026 results release

Assessment: the volume inflection is genuine and the pricing is not. A resale book earns the vintages that happen to deliver, and three quarters of 2026 have now produced realized price changes of plus 52%, plus 3% and plus 10% while published enrichment prices rose steadily throughout. The important number is the cost side, because it is the one Centrus does not control: purchase commitments under the Russian supply contract extend to 2028 and require payment on annual minimums whether or not an order is placed, and the French supply agreement runs to 2030 on a formula with floors and ceilings. A 28% increase in unit cost against a 10% increase in realized price is the reseller critique in its purest form, and it will not resolve until Piketon produces its own units.

Technical Solutions segment, and the end of the demonstration contract

Revenue of $23.7M fell 21%, with a $10.2M decline attributable to the DOE HALEU Operation Contract, and the segment swung to a $3.7M gross loss from a $3.5M gross profit. The quarterly filing explains why in a way the release does not. In June the DOE split the remaining option period into a three-month "Option 1b" requiring cascade maintenance and HALEU storage with no HALEU production, at a fixed price of $15.0M, and a 21-month "Option 1c" holding the actual production. DOE exercised Option 1b. It ended on September 30, 2026. DOE has said it does not currently intend to exercise further production options, so Option 1c, the production remainder of an option period that carried a $163.5M target cost and a $15.2M fee before the split, is not expected to be exercised.

So the segment's third quarter consisted of three months of storage and maintenance plus other contract work, and the government HALEU production business has stopped. Against the contract's cumulative obligation of $332.0M, Centrus has received $302.1M in cash. The Phase 2 fee for performance extended beyond November 2024 remains, for a fifth consecutive quarter, "not definitized and is subject to negotiation."

Assessment: the $0.8B Technical Solutions backlog should now be read as almost entirely unfunded amounts and unexercised options belonging to a customer that has stated it does not intend to exercise them. The firm performance obligations that back a revenue forecast are disclosed separately, and they are essentially all in the other segment. Last quarter we called this a new bear point; this quarter it converted from a budget proposal into a contract that ended. What makes the segment outcome better than feared is not the revenue line but the asset: the cascade did not stop with the contract.

Backlog, decomposed

The $7.0B headline is the least useful number in the release and this quarter it finally decomposes cleanly against the filings.

Assessment: two things improved materially and one is now close to worthless. The improvement nobody mentioned is the firm book: fixed-commitment performance obligations rose to roughly $0.8B and their horizon extended from 2030 to 2034, which means new long-dated utility contracts were signed during the quarter, not merely more contingent ones. That is the first durable answer to a critique this file has carried since initiation, namely that the part of the company which actually earns money had under two years of visibility. The contingent book nearly doubling to $5.4B with 94% under definitive agreement is the headline. The Technical Solutions line should now be modelled near zero.

Key Topics & Management Commentary

Overall Management Tone: the most assured posture of the five calls we have covered, and the least responsive to the share price. Management delivered three early milestone completions, a contingent backlog up 80% and a vertical-integration acquisition as a sequence of accomplishments, and deferred every quantitative follow-up to a December 7 investor day, a deferral used in answer to seven of the sixteen questions. Analyst questioning was uniformly constructive and did not reach the lease, the warrants or the capital cost, so the call's least convincing passages were the ones about commercial HALEU economics, where management declined to characterize margins at all.

1. The first centrifuge exists, and so does the supply chain behind it

The operational commitment that mattered most for 2026 was delivered two months early, and the company was careful to frame the achievement as a production line rather than a unit.

"When we announced the first centrifuge, obviously, the big announcement there is not in the centrifuge itself, but the announcement is that now we have a functioning operating production line that produces centrifuges. And we really are celebrating the first off to come off this line." — Amir Vexler, President and CEO

Alongside it, contracting with suppliers deemed critical went from roughly 75% to 100%, and the first certified-for-construction package is complete. All three were year-end targets and all three closed early. The October 29 announcement added that cumulative operating experience now exceeds four million machine hours and restated that Oak Ridge is the only advanced centrifuge manufacturing plant in the United States, with every other technology in commercial operation manufactured exclusively overseas.

Assessment: this is the quarter the technology risk became a manufacturing-rate question. For four quarters this file graded promises; this quarter three of them closed, early, and the next unknown is cadence rather than capability. Management declined to give a production rate and deferred it to December, which is a reasonable thing to withhold and a conspicuous thing to withhold twice.

2. The cascade is now enriching commercial HALEU, and Centrus paid for the privilege in liabilities

The single most consequential commercial development is stated in the release as one bullet and explained nowhere in it. On October 20 Centrus signed a transfer agreement under which the DOE moved title to materials produced or procured under the HALEU Operation Contract, including HALEU itself, low-enriched uranium feed and heels, depleted tails and the associated cylinders. In exchange, Centrus assumed the DOE's financial and operational responsibility for the facility costs, maintenance, security and environmental compliance associated with those materials, and agreed to hold the HALEU for resale to the DOE on request.

"it is our ambitions and the DOE's ambition to keep those centrifuges spinning throughout the their life cycle, and we are currently working that in through our long term arrangement with the DOE... they have given us rights to use those commercial. Those centrifuges for commercial use currently... So, currently, we are enriching for HALEU for commercial needs at this moment." — Todd Tinelli, Senior Vice President, CFO and Treasurer

Management's own framing of what that makes the company is the strongest claim on the call.

"we are excited to announce that we are now using the existing 16 centrifuge cascade to enrich HALEU for the commercial market. Making Centrus the first US owned commercial enricher for the market since our gaseous diffusion days." — Amir Vexler, President and CEO

Assessment: in substance this answers the bear point we opened last quarter, three months before the deadline we set for it. The 16-centrifuge cascade was the asset at risk when the DOE zeroed its funding, and it is now producing for commercial customers under granted rights with title to the existing inventory transferred across. What is not answered is the price. Centrus took on facility, maintenance, security and environmental obligations with no disclosed cost, and agreed to hold HALEU for resale to the DOE on request, which is an undisclosed call on its own inventory. The full amended agreement that would make this permanent remains in discussion.

3. The Piketon lease expires on October 1, 2027, and December 31 is a deadline with an accounting consequence

This is the finding of the quarter and it is in the quarterly filing published on the morning of the call. Contemporaneously with the transfer agreement, on October 19 the lease for Piketon was amended. The expiration date "was fixed at October 1, 2027," and the DOE "expressly permitted continued use of the demonstration cascade." The filing then states the condition plainly:

"The DOE has indicated that it will work diligently with the Company to complete a long-term extension of the lease for Piketon before December 31, 2026. However, the DOE is under no obligation to do so, and there can be no assurance that an extension of the Company's existing lease will be executed by December 31, 2026, or at all... In addition, if an extension of the Company's existing lease is not executed by December 31, 2026, the Company may be required under applicable accounting standards to impair or expense all or a portion of the capitalized improvements the Company has made at Piketon, which would result in a significant non-cash charge in the period in which it is recognized." — Form 10-Q for the quarter ended September 30, 2026

Net property, plant and equipment is $312.9M, against $142.5M at June 30 and $29.5M at December 31, and $60.2M of that build was still unpaid in accounts payable and accrued liabilities. The same amendment also reversed the allocation of decommissioning liability: the June 30 filing stated that "the DOE assumed all D&D liabilities arising out of the HALEU Operation Contract," while the September 30 filing states that Centrus "would assume from DOE liabilities for the decontamination and decommissioning of the facilities and equipment installed and any work performed under the HALEU Operation Contract," plus additional liabilities arising from its own use of the cascade, subject to regulatory approval of a decommissioning funding plan. No amount is attached to either item.

Assessment: the risk in this name has moved. For four quarters the unquantified variable was the capital cost of the plant. It is now the tenure of the ground the plant stands on. A company spending $350M to $500M a year to industrialize a leased federal site, with a lease that runs eleven months and a December 31 accounting trigger against a property balance that grew tenfold in nine months, has concentrated an enormous amount of thesis into one eight-week negotiation. The release discloses this only as the first bullet of its forward-looking statements, which is precisely where last quarter's funding withdrawal hid in the backlog paragraph. The pattern is now established: in this name, the material negative is in the filing, not the headline.

4. The backlog nearly doubled, and the firm book quietly gained four years

Contingent LEU and HALEU commitments rose from $3.0B to $5.4B, with $5.1B under definitive agreement against $2.4B last quarter. Management attributed roughly $2.4B of enrichment sales growth to the LEU segment and declined to split it when asked directly whether the increase came from the three newly announced advanced-reactor contracts or from undisclosed customers.

"that is not something that we disclosed, but, I can say that this was a mix of both LEU and HALEU... We do not provide the breakdown and separate the amounts. Vast majority of the backlog is LEU, either for LEU customers or a feedstock for our future HALEU offtakers." — Amir Vexler, President and CEO

The detail no document highlights is in the revenue recognition note. Firm remaining performance obligations, which exclude everything contingent, rose to roughly $0.8B and now extend to 2034, against roughly $0.7B extending to 2030 a quarter earlier.

Assessment: "a mix of both LEU and HALEU" with three advanced-reactor contracts announced in the quarter implies that undisclosed utility volume is in the increase, and the performance-obligation note corroborates it: four years of added duration on fixed commitments does not come from microreactor developers. This is the most encouraging single disclosure of the quarter and it was not in the release, not in the prepared remarks, and not asked about.

5. The September financing, and the $2 billion that is no longer there

Centrus sold 500,000 shares, pre-funded warrants over 2,005,513 shares at a nominal $0.10, and common warrants over 6,992,382 shares, for approximately $500.0M gross. The CFO's characterization was emphatic.

"This important transaction allowed us to access near term capital at a premium while setting the company up for significant future capital... If exercised in full, the warrants could provide approximately $2 billion of additional gross proceeds... We believe this structure reflects investor confidence in the long term value of Centrus while providing additional flexibility." — Todd Tinelli, Senior Vice President, CFO and Treasurer

Both halves of that deserve testing. The premium holds only if the warrants are free. The $500.0M bought 2,505,513 share-equivalents at roughly $199.6 each, about 10% above the $181.49 pricing-day close, but the company's own accounting valued the common warrants at roughly $266.8M at issuance. That leaves about $233.2M, or roughly $93 per share-equivalent, for the stock itself, about half the pricing-day close. The $2 billion does not hold either. The four warrant series are struck at 125%, 150%, 175% and 200% of that same $181.49 close, which puts them at $226.8625, $272.2350, $317.6075 and $362.9800, expiring between September 2028 and November 2031.

Assessment: the company raised $500M, paid for the headline premium with warrants, and told the market to expect $2.5B. Eight weeks later the stock is 55% below the $199.6 unit price and the second leg requires a 2.5x move at the cheapest strike. The honest accounting is that Centrus has $2.2B of cash, a signed $900M milestone-paid government contract and undisclosed customer prepayments ahead of it, which is a strong funding position and roughly four to six years of the guided spending rate. It does not have $2B of warrant capital, and any model that carries that line should remove it. The "investor confidence" framing was reasonable on September 9 and has not survived the quarter.

6. GAAP earnings are now a function of the share price

Because the common warrants are liability-classified and remeasured each reporting period, their fair value falls when the stock falls, and that decrease runs through the income statement as a gain. The CFO flagged this cleanly and introduced a new reporting basis for it.

"In the third quarter, we recognized a $46.4 million book gain due to a decrease in the liability associated with the valuation of the warrants from our recent capital raise. This was a noncash impact on our books but led to a $46.4 million positive impact to the quarter's operating income. Net income and earnings per share that is not part of our due course of business." — Todd Tinelli, Senior Vice President, CFO and Treasurer

The liability stood at $220.4M at quarter end, with the stock at $139.49. At the $89.03 reaction close every tranche is further out of the money, so a fourth-quarter gain of the same sign and material size should be expected if the price stays near here.

Assessment: management handled this correctly, naming the distortion in the first thirty seconds of its prepared financial remarks and creating a normalized basis to strip it. The disclosure is better than the market's use of it. A reader should treat the normalized loss of $(0.49) and the adjusted loss of $(0.46) as this quarter's results, note that a falling share price will keep producing reported profit, and discard any aggregated GAAP earnings estimate for this company until the warrants expire or are exercised.

7. Westerman is the first vertical-integration deal, and it carries a diversification argument

On October 16 Centrus acquired 100% of Westerman Holdings, rendered "Westermann" on the call, from its employee stock ownership trust for approximately $116.0M, paid $95.0M in cash and $21.0M in shares. The filing discloses that approximately $31.5M went to Westerman's subordinated noteholders and $12.7M to a term lender out of that price, so $44.2M of the consideration retired acquired debt. Management's case is part supply-chain insurance and part earnings.

"We view Westermann as a world class operation, especially in the nuclear space and they reach over into other industries. This allows diversity to Centrus income stream and revenue. it is immediately accretive in year 1. And it really serves as platform for us to expand on supply chain integration to take out some of those vulnerabilities if we see them present themselves through either organic growth now through Westermann or through additional M&A activities." — Todd Tinelli, Senior Vice President, CFO and Treasurer

Asked directly about gross margin and seasonality, management gave a qualitative answer across three end markets, said nuclear and high-growth segments carry stronger margins than the historical oil and gas base, and deferred the figures to December. Both the revenue and the capital guides explicitly exclude Westerman.

Assessment: the strategic logic is the most defensible acquisition rationale available to this company: it bought the maker of its centrifuge casings in the quarter it reached 100% critical-supplier contracting, from a seller whose ownership structure made the process non-competitive. "Immediately accretive in year one" is a meaningful claim from a business running a gross loss, and it is unquantified. A buyer should note that the first acquisition in this company's modern history was announced alongside an open invitation to do more, by a management team that has not disclosed the cost of its own project.

8. The Piketon capital cost is withheld for a fifth quarter, and 2027 was declined again

Asked about fourth-quarter and 2027 capital cadence, the CFO confirmed the annual guide, described the ramp, and pushed the number out.

"So we have maintained our capital spend guidance. So, obviously, you can kind of back into our fourth quarter spend What I will say is we are now crossing the threshold into deploying capital at a more accelerated rate... As for 2027, you know, what I said earlier this year in our previous calls, we are gonna get to a point where we start to see some real cadence of capital deployed. And we are now in that ramp up... So stay tuned. We will provide more guidance as for 2027..." — Todd Tinelli, Senior Vice President, CFO and Treasurer

Assessment: five consecutive quarters without a total project cost and a second consecutive quarter in which 2027 capital was explicitly declined. The December 7 investor day at Piketon is now the venue for the capital cost, the 2027 budget, the Westerman margin profile, the centrifuge production cadence, the realized portion of the cost savings and updated guidance. That is a great deal of load on one event, and it is the reason this quarter does not change our rating in either direction.

9. The cost savings programme is still identified rather than realized

Asked to split the previously announced $300M of potential savings between validated and prospective, management described the first centrifuge as evidence and claimed most of the value is ahead.

"We did announce an amount of savings that we put out there, and then you quoted $300 million, a lot of it has been a strong contributor to our capability to deliver ahead of schedule. From where I sit, and I am fairly connected to the project, I think that most of the opportunities and most of the potential savings and value we are going to derive is still ahead of us." — Amir Vexler, President and CEO

Assessment: two quarters after the figure was first published, no portion has been described as realized, no baseline has ever been given, and the answer attributes the early delivery of the first centrifuge partly to the programme without quantifying it. The claim is directionally credible because the milestone did arrive early. It remains unmodellable, and a saving against an undisclosed project cost cannot be expressed as a percentage of anything.

10. The national-security leg advanced on paper and not in dollars

Management put weight on three items: a November 2025 notice of intent from the National Nuclear Security Administration to issue a sole-source contract for certain low-enriched uranium for national security, a Government Accountability Office report describing a $37 billion federal plan to rebuild the unobligated nuclear fuel supply chain by 2052, and the October confirmation that its own supply chain is fully unobligated.

"A recent US GAO report raised the alarm on the nation's entire domestic nuclear fuel supply chain in particular, the domestic supply chain suitable to meet national security demand. Notably, the report highlighted the government's $37 billion plan to rebuild the unobligated supply chain by 2052. As announced last week, we have confirmed that our supply chain is fully unobligated. Centrus has proven, derisked technology is additionally the only production ready option for the national security establishment." — Amir Vexler, President and CEO

Assessment: the unobligated supply chain is a real competitive moat and the first tangible progress on this leg in five quarters, because an all-domestic, unencumbered component chain is the specific precondition for defence work. The sole-source intent has now been described for five consecutive quarters with no value, scope or date, and a notice of intent is not a contract. Treat this as free optionality with a non-zero probability, not as a forecast line.

11. Prepayments are described on the call and not visible on the balance sheet

Management said the X-energy, Radiant and Antares agreements "all included prepayments to support our build out," and restated prepayment as the standard structure for future HALEU offtake. Deferred revenue and advances from customers moved from $85.0M at June 30 to $86.0M at September 30, including a newly created $0.7M non-current line.

Assessment: three counterparties, three descriptions of prepayment and no disclosed amount, in a quarter where the customer-advance balance rose $1.0M. Because that balance also unwinds as ordinary deliveries are recognized, a flat total does not prove nothing arrived, and it cannot size what did. Prepayment remains the right structure and the strongest non-dilutive funding avenue available to this company, and after two quarters of being named it is still not a number.

12. The balance sheet de-levered its inventory position, and nobody noticed

Inventories fell from $377.1M to $177.7M in the quarter, inventories owed to customers and suppliers fell from $235.7M to $54.0M, and short-term inventory loans, $38.9M at year-end, were already at zero by June 30. Property rose $170.4M. Operating cash flow for the quarter was a positive $5.6M, derived against a nine-month outflow of $11.1M, alongside $131.6M of quarterly capital expenditure.

Assessment: last quarter this file noted inventories building ahead of the 2028 import prohibition and flagged the working capital as an unanswered question. The build reversed almost entirely, and most of the $199.4M decline is matched by the $181.7M fall in inventories owed to customers and suppliers rather than by sales. This is a cleaner balance sheet than the one we described in August and the change went unmentioned by the company and unasked by the room.

13. Federal support through a changing Congress

Asked whether support for domestic enrichment is insulated from the political cycle, the chief executive was unequivocal.

"I will tell you with firm belief that this ought to not change the momentum of our industry... we have now been through this cycle of 2 administrations, at least, during my time and I have not seen a change in support focus or dedication, I have seen it increase... So the answer to your question is I absolutely do not see any change." — Amir Vexler, President and CEO

Assessment: the track record supports the generic claim and this quarter contains a specific counterexample. Policy support for domestic enrichment has indeed risen across administrations, through legislation that removed the largest supplier from the US market. Appropriated funding for this company's own demonstration cascade was nonetheless withdrawn in a budget proposal, and DOE does not currently intend to exercise further production options. Support for the industry and appropriations for the issuer are different things, and the answer treated them as one.

Guidance & Outlook

The release says the company "is updating some of its financial and operational guidance," and no number moved. What changed is that three completed operational targets were lifted out of the forward guidance list and into a list of achievements, and the Q1 assumption about finalizing and funding the $900M task order is gone because it is signed. A careful reader should also note that the Piketon hiring target of 175 was raised in the August release, not this one, so the "up from 100 net new employee hires" phrasing is carried-over language rather than a second increase.

Implied fourth quarter: nine-month revenue of $308.3M against the guide leaves $141.7M to $191.7M, with a $166.7M midpoint. For scale, of the nine quarters back to Q3 2024, only Q2 2026, at $176.1M, exceeded $160M. The low end is roughly in line with the last two fourth quarters, which printed $146.2M and $151.6M. The midpoint would be the second-largest of those nine quarters and 14.0% above Q4 2025. The high end would exceed all of them.

Implied fourth-quarter capital spend: the quarterly disclosures sum to $45.2M, $82.2M and $134.0M, or $261.4M for nine months, leaving $88.6M to $238.6M. Here the two ends of the guide say opposite things. The low end would require a 34% sequential decline in spending in the quarter management described as "crossing the threshold into deploying capital at a more accelerated rate." The high end would require a 78% increase. The reaffirmed range has become wide enough to be uninformative about the only thing it was ever useful for.

Street at: the aggregated 2026 revenue estimate sits at roughly $478.6M, modestly above the $475M guide midpoint and inside the range. The aggregated earnings estimate is not usable for the reasons set out above.

Guidance style: conservative on revenue and now uninformative on capital. This management has raised guidance once, eleven weeks after first giving it, and has since reaffirmed the same range twice through quarters that varied threefold in revenue. The build milestones, by contrast, have been set and beaten: all three for 2026 landed early, and the two hiring targets were reaffirmed. The pattern is a team that under-promises on operations and declines to forecast the financials it cannot control.

Analyst Q&A Highlights

Unobligated production and the national-security opportunity

The call opened on the national-security leg, and specifically on whether the newly manufactured centrifuges carry the unencumbered status that defence work requires. The answer confirmed the status and the engagement, and supplied no size, scope or date, which has been the shape of this answer for five consecutive calls.

Q: "So first centrifuge production, that is a big milestone for the company and also national security. But a key point for us on that front was these new centrifuges are unobligated. And I was hoping you could give us some update on the engagement with the NNSA. And speaking to the funding and the outline from the GAO as well."
— Mark Shooter, William Blair

A: "a notice of intent to issue Centrus a sole source contract for the production of certain low enriched uranium that is suitable for use in support of the national security mission. So that is sort of the backdrop to our need and our drive to have unobligated production... Specifically to your question, our production will be unobligated. And we have been working to configure the supply chain, working with all of our suppliers to make sure that is going to be implemented."
— Amir Vexler, President and CEO

Assessment: the confirmation that production will be unobligated is the substantive content, and it is worth more than it sounds, because an all-domestic unencumbered component chain is a genuine barrier that no competitor manufacturing overseas can clear. A notice of intent that has gone a full year without a contract is still an option rather than a forecast.

Converting the demonstration cascade into a commercial asset

The most valuable exchange of the call came from a question about negotiations to commercialize the 16-centrifuge demonstration cascade and whether volumes could reach the market sooner. The chief executive described discussions and the chief financial officer then disclosed, in response, that the conversion has already partly happened.

Q: "can you give us an indication or kind of a status update on where you are in negotiations with the DOE and the potential to convert your demonstration cascade into commercial off take and the potential to bring some of those volumes online a little bit quicker to market."
— David, on for Jon Windham, UBS

A: "The natural next step for the... 16-centrifuge demo cascade is to place that into commercial operations. The cascade is producing. We are in discussions with the Department of Energy about how to amend and how to make a more wholesome agreement where this continues to serve the market... they have given us rights to use those commercial. Those centrifuges for commercial use currently... So, currently, we are enriching for HALEU for commercial needs at this moment."
— Amir Vexler, President and CEO, and Todd Tinelli, SVP, CFO and Treasurer

Assessment: management answered more than was asked and buried the lead in a follow-on. The rights exist, production is running and the permanent agreement does not exist yet. This is the de-risking event of the quarter on the government side, and the room moved on without asking what Centrus gave up to get it. The filings answer that question: facility, maintenance, security and environmental costs, plus decommissioning liabilities, none of them sized.

What made the centrifuge timeline accelerate, and whether it repeats

A question on the drivers of the early delivery and their transferability to serial production drew an answer about organizational incentives and process-improvement teams, a reference to the technology partnership, and a deferral on cadence.

Q: "could you just provide a little bit more color on what enabled the accelerated timeline for the first completed centrifuge? And do you see this timeline as sort of replicable for future units? And I guess the third part what learnings are you going to take from this first 1..."
— Peyton, on for Joseph Reagor, Guggenheim Securities

A: "I have been very consistent on all of our calls in trying to explain and emphasize to the investors how critically important it is for us to gain market credibility, be able to meet and exceed our commitments... we have an incentive to beat the timelines that we have advertised. And so, organizationally, we are creating the same incentives and the same expectations of exceeding and meeting the timeline... what enabled is we do run process improvement team in parallel."
— Amir Vexler, President and CEO

Assessment: a management answer about incentives in response to a question about engineering. The substance offered was that improvement teams run in parallel with production and that the learnings transfer, which is plausible and unverifiable. The specific thing an investor needs, units per month, was deferred to December for the second time in the call.

Commercial HALEU economics against the government contract

A multi-part question asked which customer the commercial HALEU serves, when revenue appears, and how margins compare with the DOE work. The first two parts went unanswered and the third was declined.

Q: "Can you just confirm whether that is against the... X-energy, Radiant, or Antares contracts, or is there another customer And then when we think about revenue and economics from that, I mean, when do you expect first commercial HALEU revenue to start showing up in the P&L And then how should we think about the margins versus the DOE HALEU contract?"
— Lawson Winder, Bank of America Securities

A: "the answer that I will have for you is it will be supplied to the commercial market. That requires HALEU and I mean, everybody has read our announcements. there is no more detail further to that... On the commercial there is a lot more that goes in. that is involved around the commercial contract terms. Obviously, we do not get into those types of economics... there is different as to who provides the feed... So the overall price, therefore, can have some variability."
— Amir Vexler, President and CEO, and Todd Tinelli, SVP, CFO and Treasurer

Assessment: the sharpest question of the call and the least informative answer. Three questions were asked and none received a response: no customer, no revenue timing, no margin characterization, not even a direction. For a company that has just declared itself the first US-owned commercial enricher in a generation, declining to say whether commercial HALEU earns more or less than a cost-plus government demonstration contract is a conspicuous silence, and it is the single largest gap this report cannot close.

Supply-chain strategy after the first acquisition

Questioning turned to whether vertical integration is now the standing approach to supplier risk or was a one-off opportunity. Management framed the deal as opportunistic rather than programmatic, then described a platform for more.

Q: "So as it relates to your supply chain strategy, should we expect to see vulnerable areas addressed through vertical integration like with Westermann, or was that more of a sort of a 1-off event from a strategic perspective?"
— Luke, on for Eric Stine, Craig-Hallum

A: "We have a lot of suppliers Each supplier has their own strengths and weaknesses... we have built resilience in our supply chain where possible. We are using quality oversight. We are using partnership with different suppliers. So we are not taking anything off the table I think what you saw with Westermann is an example where you had a really perfect fit of all these factors come together. And it was no coincidence that this deal was made."
— Amir Vexler, President and CEO

Assessment: "not taking anything off the table" alongside a chief financial officer describing a platform for "additional M&A activities" is an acquisition programme, not a one-off. That is strategically coherent for a company that needs its suppliers more than its suppliers need it. It is also a new use of capital, announced by a management team that will not disclose the cost of the project the capital is for, and it deserves a capital-allocation framework at the December event.

The fourth-quarter bridge and whether gross margin stays negative

One analyst did the arithmetic on the reaffirmed guide out loud and asked directly whether the fourth quarter carries another negative gross margin. The answer reaffirmed confidence in the full year and did not address the margin question.

Q: "you look at the annual outlook, it looks like Q4 will be $142 million to $192 million revenues. Like, are there any takes and puts on this? and then part 2 of that question is, the SWU pricing versus SWU cost management. How are you looking at it? Should we expect another sort of negative gross margin quarter this quarter?"
— Sameer Joshi, H.C. Wainwright

A: "As you correct, we have maintained our guidance. We are optimistic and we feel obviously that the fourth quarter provides a full year view of our strong results. As we have always noted that we see significant seasonality, including when we actually have sales that come that are being delivered, but also receive our shipments. That quarter over quarter mix provides some flexibility in our gross margin profile. that is why we have been discussing this on the TTM basis to provide a more normalized view."
— Todd Tinelli, SVP, CFO and Treasurer

Assessment: the arithmetic was confirmed and the margin question was not answered. Confirming an implied $141.7M to $191.7M fourth quarter after a $55.5M third quarter is a meaningful commitment, because the midpoint requires the second-largest quarter of the nine back to Q3 2024, although the low end needs only a typical fourth quarter. The redirection to a trailing twelve-month frame is the same move made in May and August, and it is defensible for a business with annual delivery commitments. It is also an answer that cannot be falsified until the full year prints.

Capital cadence into the fourth quarter and 2027

A question on spending shape drew the clearest statement yet that the ramp is now underway and a second deferral of the 2027 number.

Q: "Hoping to get any additional clarity you guys can provide in terms of kind of Q4 spending Obviously, it could be a wide range of outcomes and still fit the, the full year guide. So just kind of wondering how we should think about spending into Q4 and any early thoughts on spending trajectory into 2027?"
— Jeffrey Grampp, Northland Capital Markets

A: "What I will say is we are now crossing the threshold into deploying capital at a more accelerated rate. We have started our first construction package As you can see, we have our first centrifuge rolling off the line. We have secured our supply chain and have orders in for all of our materials currently placed... And so now as that is beginning to come to fruition and we have boots on the ground, our spend increases."
— Todd Tinelli, SVP, CFO and Treasurer

Assessment: the analyst identified the problem precisely, that the reaffirmed range admits opposite outcomes, and the answer confirmed acceleration without reconciling it to a low end that would require a sharp decline. The qualitative content is useful: materials ordered, construction package started, engineering contractor on site. The quantitative content is that the fourth quarter will be above $134M, which makes the bottom of the capital guide unreachable and the guide itself stale.

Backlog shape and the new large-reactor pipeline

Two separate questions probed whether the backlog is weighted toward utilities or advanced reactors and whether it is front- or back-loaded. Management declined the split and gave an unusually direct statement of commercial priority.

Q: "Is there a preferred mix of I guess, utilities, say, LEU versus newer companies, SMR, and, I guess, with the shape of the backlog, you talked about backlog going up to 2040. Is that more front loaded or back loaded? I am just trying to get a sense of what that can mean for the implication on CapEx as you, I guess, continue to grow your backlog."
— Bill Peterson, JPMorgan

A: "we typically do not break down and separate the amounts That said, the vast majority of our backlog is LEU... The established LEU customers provide a lot of certainty. They provide a lot of volume. They provide very, very solid terms to be able to offtake that is precisely what we need when we are expanding our facilities and investing capital. However, I do not want to, in any way, underestimate the HALEU customers who, just in the last 6 months to a year, have become a lot more committed to the supply chain and we are seeing them being able to make sort of take-or-pay-type contract and legally binding contracts."
— Amir Vexler, President and CEO

Assessment: the honest framing of a company that needs investment-grade counterparties to finance a plant and advanced reactors to justify its premium. "Take-or-pay-type contract and legally binding contracts" from microreactor developers is new language and consistent with the three contracts signed in the quarter. The refusal to split a backlog that just grew $2.4B remains the largest modelling obstacle in the name, and the performance-obligation note in the filing does more to resolve it than the answer did.

Tails assay, overfeeding, and whether enrichment keeps the upper hand

A question on how customers optimize between uranium and enrichment as relative prices move produced the most analytically interesting answer of the call.

Q: "With the strong SWU pricing relative to uranium, as we look ahead, how are customers changing tails assay, if at all, and how much Western SWU demand could be met with overfeeding as we look out in the future?"
— Nate Pendleton, on for Jed Dorsheimer, Texas Capital

A: "Obviously, that dynamic has changed now that SWU prices have gone up considerably... I would tend to give the upper hand here to enrichment, By that, I mean, I think, ultimately, you would see the seesaw and pendulum swing towards sort of erring on the side, of favoring enrichment capacity is just because I do not see that there is that much ability to expand capacity as quickly as the market grows. I mean, we are we are seeing that disparity now, and I think it will it will take a while before that disparity sort of settles."
— Amir Vexler, President and CEO

Assessment: the structural argument for the whole investment, stated in operating terms rather than marketing terms. If enrichment capacity cannot expand as fast as reactor demand grows, utilities optimize toward buying separative work and away from overfeeding with cheap uranium, which supports both volume and price for whoever owns capacity in 2029. It is also the answer that best explains why the realized price in the current book is irrelevant to the thesis: the pricing power arrives with the new plant, not with the resale contracts.

What They're NOT Saying

1. The Piketon lease, anywhere outside the filings. The expiration was fixed at October 1, 2027 by an October 19 amendment, a long-term extension is targeted for December 31, 2026 with no DOE obligation to deliver it, and a miss may force impairment of capitalized improvements against $312.9M of net property. The release carries it only as the first bullet of its forward-looking statements. It was not in the prepared remarks and no analyst raised it.
2. What Centrus gave up for commercial cascade rights. The company assumed the DOE's responsibility for facility costs, maintenance, security and environmental compliance on the transferred materials, and agreed to assume decontamination and decommissioning liabilities the DOE held as of the June filing. No amount, no annual run rate, no reserve.
3. The value of the transferred HALEU inventory. Title to HALEU, low-enriched uranium feed and heels, depleted tails and cylinders moved to Centrus on October 20. No quantity, no carrying value, no consideration, and an undisclosed obligation to hold the HALEU for resale to the DOE on request.
4. Commercial HALEU economics. Asked for customer, revenue timing and margin against the DOE contract. Three questions, three non-answers, not even a direction on margin.
5. The Piketon capital cost and the hurdle SWU price. Fifth consecutive quarter, now alongside a signed $900M award, a disclosed $900M construction contract cap, four offtake counterparties, 100% supplier contracting and the first completed centrifuge.
6. 2027 capital spending. Declined for a second consecutive quarter, in the quarter the company said it is crossing into accelerated deployment.
7. Centrifuge production cadence. Asked twice, deferred twice to December. "A certain cadence and rhythm that we are currently engaged in" is the entire disclosure on the output of a production line backed by a $560M announced investment.
8. Any prepayment amount. Three HALEU agreements described as including prepayments, zero figures, and a customer-advance balance that rose $1.0M in the quarter.
9. Westerman's financials. "Immediately accretive in year one" with no revenue, no margin, no seasonality and no purchase accounting, excluded from both guides, deferred to December.
10. The realized portion of the $300M of cost savings. Second consecutive quarter asked directly, second consecutive quarter with no realized figure and no baseline.
11. The Phase 2 fee. Fifth consecutive quarter undefinitized and subject to negotiation, on performance completed in June 2025, with no analyst asking in five calls.
12. The warrant structure. Not in the release, not in the prepared remarks beyond the proceeds figure, and never raised in Q&A. Eight tranches over 6,992,382 shares struck 155% to 308% above the market, framed on the call as "approximately $2 billion of additional gross proceeds."
13. The inventory unwind. Inventories fell $199.4M and inventories owed to customers and suppliers fell $181.7M to $54.0M. Neither document mentions it and nobody asked, one quarter after the build was described as positioning for the 2028 prohibition.

Market Reaction

• Pre-print setup: LEU closed at $92.58 on November 4, down 61.9% year to date from $242.76 against the S&P 500's +12.3%, down 71.7% over the trailing twelve months from $327.20, and down 34.5% over the trailing thirty days from $141.44. The pre-print close was the low end of a 52-week closing range of $92.58 to $342.99, so the stock entered the print at its twelve-month low.
• Reaction session: the stock opened at $95.49 on November 5, 3.1% above the prior close, traded a range of $84.86 to $95.49, and closed at $89.03, down 3.8% or $3.55. The low marked an 8.3% decline and the high was the open.
• Volume: 1.5M shares against a 30-day average of 0.8M, or 1.8x normal, the second-heaviest of the five reaction sessions in this coverage.
• Market context: the S&P 500 fell 4.4% on the session. LEU outperformed the index by roughly 60 basis points.
• Peer reaction: on the same session the uranium exchange-traded fund fell 5.3%, the largest producer fell 4.7%, and the advanced-reactor cohort fell between 9.9% and 14.8%. LEU posted the smallest decline of the seven names and the index.

This is the first reaction session in this coverage where the company-specific signal is close to zero. A 3.8% decline on a day the index fell 4.4% and every nuclear comparable fell between 4.7% and 14.8% is not a verdict on the print. It is a stock being carried by its sector on a broad risk-off session.

The five-week de-rating, not the session, is the thing to explain. The stock fell from $139.49 on September 30 to $92.58 on November 4, a 33.6% decline across 24 sessions with no single gap day, falling on seventeen of them. Over the identical window the advanced-reactor cohort fell between 27.9% and 40.2% and the uranium producers fell between 17.3% and 17.9%. LEU sits in the middle of the reactor cohort and well below the producers, which locates it precisely: the market is treating Centrus as a long-duration advanced-nuclear development story rather than as a uranium-linked operating business.

The attribution is macro and consistent across sources. The September inflation release pushed markets to price a resumption of monetary tightening, and rising long yields hit pre-revenue, long-dated power names hardest. Alongside that, the demand narrative weakened at the margin: hyperscale buyers have been prioritizing near-term megawatts over long-dated gigawatt nuclear, and utilities have been slow to sign despite uranium spot breaking a nineteen-year record near $96 per pound. A sector where the commodity makes new highs while the equities make new lows is a sector being repriced on cost of capital and time to cash, not on end demand.

The uncomfortable arithmetic. Across five prints the reaction sessions have been (14.7)%, (20.7)%, +12.3%, (5.1)% and (3.8)%. The stock is down 72.7% from the $325.73 close that preceded our initiation. Over those same five quarters Centrus signed a $900M government contract, removed the financing contingency from more than $3B of customer commitments, named four HALEU offtake counterparties, contracted 100% of its critical suppliers, manufactured its first centrifuge two months early, became the first US-owned commercial enricher in a generation, and grew its contingent enrichment backlog from $2.3B to $5.4B. The share price and the asset have now moved in opposite directions for a full year, and this quarter the gap reached the point where the equity is worth less than the cash on the balance sheet.

Street Perspective

Debate: Is a market capitalization below gross cash an opportunity or a message?

Bull view: the bull case treats this as the clearest mispricing in the sector. At $89.03 the fully loaded equity is worth roughly $2.00B against $2,236.1M of unrestricted cash, so after netting $1,178.8M of debt the market is assigning roughly $0.94B of enterprise value to a company with a $5.4B contingent backlog that is 94% definitive, a signed $900M government contract, the only US advanced centrifuge manufacturing plant, and a technology with over four million machine hours. That is two times trailing revenue of $454.5M for an asset with a statutory supply cliff arriving fourteen months out.

Bear view: the bear response is that cash which is contractually committed to a multi-billion-dollar construction programme is not cash available to shareholders, it is a construction budget with a 2029 completion date and no disclosed total. The current business runs a gross loss, consumed $237.5M before financing over nine months, and depends on supply contracts whose unit costs rise faster than its realized prices. The market is not mispricing the balance sheet. It is discounting a four-year cash burn into an undisclosed capital cost on leased ground.

Our take: the bears have the better description of the mechanism and the bulls have the better description of the price. Both are right that the cash is a budget rather than a return, and the question is whether the budget buys an asset worth more than it costs. That cannot be answered until the capital cost is disclosed, which is the fifth quarter we have written that sentence. What is new is that the market has stopped paying anything at all for the outcome: at 0.17x a contingent backlog whose only remaining conditions are operational milestones, the enrichment option is now close to free, and the downside is bounded by a balance sheet rather than by a narrative. That asymmetry is why the rating holds.

Debate: Does the commercial cascade conversion offset the loss of government funding?

Bull view: the bull argument is that the conversion is strictly better than the contract it replaces. The demonstration contract was cost-plus work on DOE-owned output at a target fee of roughly 9%, and it was always transitional. What replaces it is title to the output, the right to operate the only HALEU cascade in the Western world for commercial customers, and four named offtake counterparties, three of them under contract, to sell into. Fee-based government revenue became a merchant position in a market with no other supplier.

Bear view: the bear response is that the segment swung to a $3.7M gross loss, the last exercised option ended on September 30, roughly $0.8B of backlog belongs to a customer that has said it does not currently intend to exercise further production options, and the company accepted facility, maintenance, security, environmental and decommissioning obligations of undisclosed size to get the rights. Management then declined to characterize commercial HALEU margins in any direction. An unsized liability traded for an unsized revenue stream is not demonstrably an improvement.

Our take: the bulls are directionally right and the bears correctly identify that nobody can prove it. Owning the output of the only operating HALEU cascade in the West, with three contracted counterparties and a fourth under a letter of intent, is a better asset than being paid a fee to run it for its owner. The refusal to say whether it earns more or less than the contract it replaces is the most conspicuous non-answer of the call, and it means the offset is a judgment rather than a calculation. We grade the bear point we opened in August as answered in substance and unquantified in fact, and we will not call it closed until the permanent agreement is signed and its economics are described.

Debate: Was the September financing a success or a warning?

Bull view: the bull reading is that this was an exceptionally well-executed raise. The company sold units at a headline price roughly 10% above the market, took in $493.8M net, left its at-the-market programme untouched, and attached warrants that only convert into dilution if the stock trades 25% to 100% above the $181.49 pricing-day close. Shareholders got $500M without a headline discount and gave up only upside they would be delighted to pay for.

Bear view: the bear reading is that the premium was purchased with the warrants, not earned, and that the market paid $199.6 for something now worth $89.03. The structure also installed a $220.4M liability that makes reported earnings move inversely to the share price, destroyed the comparability of the earnings consensus, and was presented to investors as the first instalment of $2.5B when the remaining $2B now requires at least a 2.5x move. Common warrants over 34% of the outstanding share count sit above the market with up to five years to run.

Our take: both are true and the sequence matters. The raise was well timed on September 9, its headline premium was paid for with warrants, and the characterization of its second leg has not survived contact with the share price. The capital that exists is real, large and sufficient: $2.2B against a guided spending rate of $350M to $500M a year is four to six years of runway, and the first cascade arrives in 2029. The capital that was advertised is gone, and the right response is to delete the $2B from the funding plan rather than to argue about whether it was ever reasonable to include. We hold management to the first standard and mark down the second.

Model Update Needed

Valuation: at the $89.03 reaction close, on 20,453,108 shares of Class A and Class B common stock outstanding at September 30 plus 2,005,513 pre-funded warrants that the company itself includes in basic earnings per share, the fully loaded equity is worth roughly $2.00B. Netting approximately $1,057M of net cash gives an enterprise value near $0.94B. That is 2.1x trailing-twelve-month revenue of $454.5M, 2.0x the midpoint of the 2026 guide, 8.9x trailing gross profit of $105.8M, 12.2x trailing adjusted net income of $77.3M, 0.17x the $5.4B contingent enrichment backlog and 0.13x the $7.0B total. Enterprise value has fallen 67% since our August note and roughly 80% since initiation, while the contingent backlog has grown from $2.3B to $5.4B and the definitive share of it from 91% to 94%.

Two framings are worth stating plainly. First, the equity is worth 89% of the company's gross cash balance, or roughly 93% on a pro forma basis after October's $95.0M Westerman payment. Gross cash per fully loaded share is $99.57 against an $89.03 price. Second, the enterprise value is now roughly equal to the $900M cap on a single construction contract the company has signed, and to the $900M government award that funds part of the work.

We maintain Outperform and cut the fair value range to $140 to $185 from $240 to $300. The $140 low end implies +57.3% from the $89.03 reaction close, the $185 high end +107.8%, and the $162.50 midpoint +82.5%. At the midpoint the enterprise value would be roughly $2.59B, or 0.37x the $7.0B total backlog and 5.7x trailing revenue. In August the market price itself carried 0.64x a $4.5B total backlog and 6.0x trailing revenue, and our $270 midpoint carried roughly 1.04x and 9.9x. In other words our new midpoint applies roughly a 65% multiple haircut to a total backlog 56% larger and more definitive, and sits below the multiple the market itself paid in August.

The cut is 40% at the midpoint and we want to be exact about why, because none of it is a changed view of the asset. Roughly half reflects the sector's cost of capital: a rate-driven repricing has taken the whole advanced-nuclear complex down 28% to 40% in five weeks, and a story whose first cash flows arrive in 2029 is correctly worth less when the discount rate rises. The remainder reflects two specific facts from this quarter. The $2B warrant leg of the funding plan is unreachable at this price and should be removed from any model that carried it. And the Piketon lease now carries a dated, eight-week deadline with an impairment consequence against a property balance that grew tenfold in nine months.

We would note plainly that an implied return above 80% is well above what our Outperform band normally carries, and that this is the second consecutive quarter in which the reason is a falling share price rather than a rising valuation. A reader is entitled to ask whether an analyst who has maintained Outperform through a 72.7% drawdown is describing an opportunity or defending a position. Our answer is that the test should be the asset, and on the asset this was the best quarter of the five: all three 2026 build milestones met early, the contingent backlog nearly doubled, the firm book extended four years, and the bear point we opened in August answered in substance. The honest caveat is that none of that has mattered to the share price for a year, and a holder must now be willing to sit through a 2028 supply cliff, a 2029 first cascade and an undisclosed capital cost to collect.

Thesis Scorecard Post-Earnings

Five items were set out in August to be graded this quarter.

Overall: the best operational quarter and the worst financial quarter of this coverage, in the same release. All three 2026 build milestones landed early, with the two hiring targets still open, the contingent backlog nearly doubled with its definitive share rising to 94%, the firm order book gained four years of duration, and the bear point we opened in August was answered in substance three months before our own deadline. Against that, the gross loss and the operating loss are the deepest of the nine quarters back to Q3 2024, the price-cost spread was adverse for a second consecutive quarter and widened, the advertised second leg of the funding plan evaporated, and a new bear point opened on the capital structure. The thesis is unchanged in direction and lower in value.

The single most important thing in the quarter is not in the earnings release. The risk in this name has migrated from the cost of the plant to the tenure of the ground beneath it: a lease fixed to October 1, 2027, a December 31 extension deadline that the counterparty is under no obligation to meet, an impairment trigger against $312.9M of capitalized improvements, and decommissioning liabilities that Centrus has agreed, subject to regulatory approval, to take from the DOE without a number attached. For the second consecutive quarter the most material negative disclosure sat outside the release bullets, outside the prepared remarks, and outside the sixteen questions asked on the call.

Action: maintaining Outperform; cutting fair value to $140 to $185 from $240 to $300. On the August trigger requiring a commercial cascade agreement by the Q4 print, the substance arrived early and the permanent agreement has not, so the trigger remains live but satisfied in direction. On the August trigger requiring a Piketon capital cost at the December investor day, nothing has changed and the test is 32 days away. We add one trigger and make it the sharpest in the file: we revert to Hold if the Piketon lease extension is not executed by the year-end deadline, because an impairment of the capitalized build would convert a disclosure problem into a reported loss and a renegotiation into a hostage situation. What we will grade next quarter and at the December 7 event: whether the lease extension is executed; whether the Piketon capital cost is finally disclosed; whether 2027 capital spending is guided; whether the implied $141.7M to $191.7M fourth quarter arrives; whether commercial HALEU economics are described in any direction; and whether any offtake produces a disclosed prepayment amount.