Anthropic's IPO Valuation Collapses to 1.3 Trillion as Investors Demand Proof of Profitability

The most anticipated technology IPO in a generation is repricing in real time, and the direction is down. Prospective investors in Anthropic's planned public offering are now discussing a valuation of $1.2 trillion to $1.3 trillion, a drop of $200 billion to $300 billion from the $1.4 trillion to $1.5 trillion range communicated to bankers just days prior, and less than two-thirds of the $2 trillion figure that circulated in early October, according to Bloomberg reporting cited by Investing.com.

The revision did not happen in isolation. On Thursday, the Nasdaq 100 posted its largest single-day decline on record, falling 6.4% and erasing roughly $2.3 trillion in market value. The Cboe Volatility Index climbed to its highest reading since April 2025. The Wall Street Journal reported that three major prospective investors indicated the offering would need to come at a lower price, with one pushing for a valuation closer to $1 trillion.

The roadshow, led by Goldman Sachs and Morgan Stanley alongside 20 other banks, was targeting a launch the week of November 9, with pricing the following week, according to Reuters. But four people familiar with the discussions told Reuters that investors are questioning whether the company can hold that timeline, partly because Anthropic left its October revenue figure blank in an updated S-1 filing.

The gap between where Anthropic wants to price and where investors will accept is not a minor rounding error. It is a referendum on how public markets value frontier AI at a moment when the sector's cost structure and revenue returns are both under scrutiny.

Three Steps Down: How Valuation Expectations Retreated

Anthropic's valuation compression follows a pattern that is now measurable in three discrete stages. The first discussions in early October reached $2 trillion. After investors received a briefing on financial performance, the range fell to $1.4 trillion to $1.5 trillion. The Nasdaq rout produced the third leg down, to $1.2 trillion to $1.3 trillion.

That trajectory, three downward revisions across roughly six weeks, is not noise. It reflects a systematic reassessment of how much growth premium the market will assign to a company whose capital requirements are structurally enormous and whose near-term revenue is not fully disclosed.

The offering aims to raise at least as much as SpaceX's record $86.2 billion debut. That target, combined with the valuation decline, compresses the implied float percentage and raises the execution risk of hitting the fundraise target simultaneously with market conditions tightening.

The stock is slated to trade on Nasdaq under the ticker ANT.

The Capital Intensity Problem That Investors Cannot Ignore

Anthropic's amended S-1 filing makes the capital structure legible and the numbers are substantial. The company has arranged significant multi-year compute commitments, and those commitments are precisely why it needs to raise capital at scale through a public offering.

The broader infrastructure context amplifies the concern. Amazon, Microsoft, Alphabet, and Meta have collectively committed to at least $720 billion in capital expenditures this year, the majority directed at AI infrastructure, with over $1 trillion in multi-year spending commitments arranged beyond that. [Supporting Source] Despite that spending, none of the four have demonstrated AI-driven monetization that covers their infrastructure outlays. Alphabet's earnings call provided the starkest illustration: AI Mode queries in Google Search grew revenue by 12% in the third quarter, down from 17% growth in the same quarter the prior year, as AI-enabled search cannibalized the core business before monetization was solved. [Supporting Source]

Anthropic sits downstream of that dynamic. It is a consumer of the same infrastructure those companies are building, and it is going public into a market where investors are already questioning whether the buildout returns justify the spend.

Caption: Anthropic IPO valuation revisions across three distinct stages. Source: Bloomberg via Investing.com, Wall Street Journal .

The METR Capability Report: A Double-Edged Asset

One of the more unusual elements of Anthropic's S-1 process is the inclusion of an independent capability evaluation by Model Evaluation and Threat Research, known as METR. The firm received what it described as "unprecedented access to internal data" over roughly eleven weeks, from mid-August through late October.

METR's finding: Anthropic is improving model capabilities on well-specified tasks at approximately twice the rate it was a year ago. METR said it could rule out compute growth as the primary driver, and attributed the acceleration to AI systems assisting in the development of newer AI systems, what Anthropic's own filing describes as "closing the loop."

The bull case embedded in that finding is obvious. A company doubling its capability improvement rate, driven by recursive AI-assisted development, is compounding faster than its hardware spend would suggest.

The caveats are equally significant. METR noted the acceleration may apply only to tasks that can be automatically graded, not to unstructured real-world applications. The team also flagged internal disagreements and acknowledged the results could reflect understatement of prior progress. Anthropic did not respond to requests for comment on the report.

The irony runs deeper. CEO Dario Amodei published an essay on September 12 calling on the industry to slow the pace of AI capability improvement, warning that recursive self-improvement could outrun humanity's ability to understand and control the systems. The METR report, commissioned under a process Amodei championed, now sits inside an IPO filing that is asking public investors to pay a trillion-dollar premium for the speed of that exact same capability curve.

Infrastructure Stress Fractures Are Spreading

The Anthropic repricing did not arrive in a vacuum. Across the AI infrastructure complex, credit and equity markets are signaling distress simultaneously.

Microsoft tightened oversight of data center projects from neocloud providers Nebius, IREN, and Nscale, Bloomberg reported Thursday. Nebius, IREN, and CoreWeave have each lost roughly half their market value over the past month. Credit default swaps on CoreWeave and SpaceX hit record levels, implying rising default probabilities among investors pricing those instruments.

Technology stocks underperformed the S&P 500 by 10.7% over the past month, the most extreme one-month underperformance since May 2002, according to Truist Advisory Services. Vanda Research characterized retail selling during the period as the strongest since the COVID-19 pandemic.

Ballot measures in multiple states that would limit data center development have added a political dimension to what was previously a capital allocation question. The regulatory overhang on AI infrastructure, once theoretical, is now crystallizing in ways that affect permitting timelines, power access, and long-term capacity assumptions for every model developer dependent on that infrastructure, including Anthropic.

Investment Positioning: Where the Risk-Reward Sits

For institutional capital evaluating the Anthropic IPO at a $1.2 trillion to $1.3 trillion range, the question is not whether the company is building something consequential. The question is whether the entry multiple prices in both the upside scenario (recursive capability acceleration, dominant enterprise AI adoption) and the downside scenario (delayed monetization, infrastructure bottlenecks, regulatory restrictions on data centers, and a public market that now demands nearer-term earnings visibility).

Bloomberg's reporting indicates investors and bankers are still working toward an offering rather than abandoning the timetable. That matters. A postponement carries its own signal cost for the broader AI ecosystem, potentially confirming that the sector cannot sustain public market scrutiny at the valuations private rounds established. A deal that closes, even at a haircut, preserves optionality.

The fundraise target, at least matching SpaceX's $86.2 billion record debut, requires investor conviction across a broad syndicate at a moment when the 20-bank underwriting group is navigating simultaneous equity volatility, sector underperformance, and an October revenue line that remains blank in the public filing.

The Plocamium View

The three-stage valuation compression is not primarily about Anthropic. It is a price discovery event for the entire frontier AI asset class, and the market is delivering a verdict: private round valuations, set in a low-volatility environment with strategic investors who had non-financial motivations for participation, do not translate cleanly into public market multiples.

What the market is implicitly demanding is a bridge between capability metrics and monetization timelines. METR's finding of doubled capability improvement rates is analytically interesting but commercially incomplete. Institutional investors at scale need a denominator. Without a disclosed October revenue figure, the valuation exercise becomes a comparables game played against companies like Alphabet and Microsoft that are themselves struggling to convert infrastructure spend into AI-driven earnings growth.

The second-order implication is for the private AI funding ecosystem. If Anthropic prices at or below $1.3 trillion after reaching $2 trillion in private discussions, every late-stage AI company that has raised capital against an inflated comparable will face its own repricing event, either at the next fundraise or at eventual IPO. The pipeline of frontier AI companies seeking public exits is not short, and each one will now price against whatever Anthropic establishes as the reference transaction.

There is also a structural observation worth making about Goldman Sachs's positioning here. Goldman elevated 109 executives to partner in November 2026, its largest class since 2010 and 15% larger than the 2024 class, with Asset and Wealth Management accounting for 30% of promotions, a record for that division. [Supporting Source: Business Insider]

The company that publicly asked the industry to slow down is now under pressure to close the fastest possible path to public markets. That tension, between Amodei's September 12 essay and the November roadshow, is not a contradiction to dismiss. It is the defining tension of frontier AI as an asset class: the people who understand the risk most clearly are also the ones most incentivized to accelerate.

Plocamium's position: a deal that prices at $1.2 trillion or below, combined with a disclosed October revenue figure that supports the trajectory implied by the METR acceleration data, creates a defensible entry point for long-duration institutional capital. A deal that prices without that disclosure, or that postpones entirely, should be read as a signal that the AI infrastructure cycle is entering a more turbulent phase than the equity markets have yet fully discounted.

The Bottom Line

Anthropic's IPO has become the most visible stress test of AI valuations in public markets. Three downward revisions across six weeks, a missing October revenue figure, record Nasdaq volatility, and credit stress across the neocloud infrastructure complex are all converging on a single pricing event. If the deal closes near $1.2 trillion, it sets a new reference multiple for every frontier AI company that follows. If it delays, the repricing will travel upstream into private markets. Either outcome reshapes institutional capital allocation toward AI for the next twelve to eighteen months.