← sources

What fund managers wrote about Anthropic in their Q3 2026 letters

source: reddit (reddit:ValueInvesting)published 2026-10-27 21:08ingested 2026-10-28 02:39

retained text

On Monday, October 26, Anthropic published its filing for a Wall Street IPO.

Several fund managers wrote about Anthropic in their Q3 2026 letters. Below is what 5 of them wrote, to help you form your own view.



Greenlight Capital: 

>"The real exercise, in our opinion, is to invert the question. How can investors have confidence in the terminal value of the large AI providers? These companies are investing trillions of dollars in aggregate, chasing highly uncertain returns, and the market is rewarding them with generous (or extremely generous) valuations. Anthropic is expected to be the next company to IPO – with a whispered valuation of $2 trillion. In 10 years, we have a strong opinion about what ACHC, CNR, FLR and GRBK will generally look like. We have no idea what Anthropic will look like or whether it will even exist 10 years from now."

>"After SpaceX (SPCX) proclaimed its total addressable market (TAM) to be $28.5 trillion, Anthropic topped it by claiming a TAM of over $30 trillion. Of course, Elon Musk responded by saying he could envision SPCX being worth more than the current economy of Earth. To put the scale of these claimed addressable markets into perspective, the 191 technology companies in the S&P 1500, plus Amazon, Meta, Google and Tesla, collectively generated $4 trillion in actual revenue last year."



1492 Capital Management: 

>"Looking ahead, Anthropic is forecast to IPO in November at a valuation approaching $2 trillion, which is a similar valuation to SpaceX. OpenAI is projected to have its IPO in the first half of 2027, and that deal is currently projected to command a valuation of around $1.2 trillion. The total of these three mega-IPO valuations is more than the first day valuation of all tech IPOs from 1980 to 2025 (3,365 IPOs in all)."

>"Anthropic CEO Dario Amodei published an essay calling for a coordinated effort to regulate and slow frontier AI, and OpenAI's Sam Altman and xAI's Elon Musk subsequently joined the calls for regulation. A leaked IPO prospectus reportedly told investors that the company's models could pose a 'catastrophic or existential risk to humanity' and can 'resist shutdown'. We take safety seriously, and we do not doubt that many researchers are sincere. However, we believe the idea that runaway AI agents will end humanity is a false narrative, and that investors should understand why the people promoting it are saying it."

>"Liability. If you tell the public your product carries a meaningful chance of catastrophe, the next logical question is for the government to absorb the legal consequences. The labs have reportedly sought precisely that, and Treasury Secretary Bessent rejected a liability exemption for AI labs at a House hearing on September 15. As he put it days later, 'it is humans who are responsible, not the AI.' Our view is that if you can build Frankenstein, you should be able to build a Frankenstein destroyer. A company cannot say its product carries a one-in-ten chance of wiping out humanity and then ask Washington to take the liability off its hands. If the labs truly believed the risk, they could slow down on their own, and nothing in law prevents it. How can these companies who want to come to the public markets through the IPO process expect the public markets to underwrite unlimited product liability exposure, yet they want to command a $2 trillion valuation in the public markets? This issue is why Anthropic's CEO has invited the U.S. government to take an investment stake of up to 50% of the company. What could be better than having the government as your partner to block competition and insulate against any significant product liability?"



Osprey Private Wealth: 

>"This negative narrative began to take center stage in mid-September as a 'whistleblower' from Anthropic, one of the two leading AI frontier lab companies, announced his resignation due to his concerns that AI development is moving too quickly without adequate safety measures, and that in his opinion, there was a greater than 10% risk of an existential threat to humanity by the end of the decade. These concerns intensified after AI models from both Anthropic and OpenAI (the other leading AI frontier lab) were found to have hacked into other company websites and demonstrated malicious behavior. In response, some in the industry and in the halls of Congress are strongly advocating for tight regulation of AI companies, while others such as Nvidia's CEO Jensen Huang dismiss the 'end of humanity' claims as not grounded in science. Huang suggests that if these leading AI companies are truly worried about safety, they can slow the release of the newer versions of their AI models themselves without the need for government regulation compelling them to do so."

>"Some observers strongly believe the regulatory angle here is a stealth move by the leading AI labs to compel the government to establish highly competitive barriers (through regulation) that would make it difficult for 'open model' companies to survive. (Both Anthropic and OpenAI are proprietary, 'closed model' companies.) This can be viewed in a similar light to the regulations that were created after the financial crisis of 2008, which ultimately made the largest banks even larger and effectively 'too big to fail'."



Montaka Global Investments: 

>"For example, in a case where Claude Mythos Preview cheated on a training task, Anthropic researchers were able to reveal that Claude was internally 'thinking' about how to avoid detection, despite omitting this from its documented chain-of-thought."

>"The regulatory and legal environment surrounding AI will likely become more restrictive. In the weeks and months leading up to the multi-trillion dollar IPOs of Anthropic and OpenAI, it is becoming increasingly difficult to see how the creation of uncontrollable AI agents that pose real cyber risks is a sustainable business model. We believe new economic and criminal liability laws will be increasingly demanded by voters. As JPMorgan's Jamie Dimon recently pointed out, allowing public access to Anthropic's Claude Mythos AI model is like giving 'ballistic missiles to individuals'. Such concerns are likely to strengthen calls for greater regulation and oversight."



Baillie Gifford US Equity Growth: 

>"As investors in Anthropic and OpenAI through our private company portfolios and in Alphabet, Meta, Amazon and other holdings in this portfolio, we welcome greater independent scrutiny and take seriously the responsibility of the companies developing the technology."



Sources:

Greenlight Capital, Q3 2026 letter: [https://www.hfbestideas.com/letters?open=o0P2WUUHgMHi](https://www.hfbestideas.com/letters?open=o0P2WUUHgMHi)

1492 Capital Management, Q3 2026 Vantage Point newsletter: [https://www.hfbestideas.com/letters?open=zO5l6VNn2Kw4](https://www.hfbestideas.com/letters?open=zO5l6VNn2Kw4)

Osprey Private Wealth, Q3 2026 letter: [https://www.hfbestideas.com/letters?open=TmRAb7zkykE5](https://www.hfbestideas.com/letters?open=TmRAb7zkykE5)

Montaka Global Investments, Q3 2026 letter: [https://www.hfbestideas.com/letters?open=fqAKPl0ClFfx](https://www.hfbestideas.com/letters?open=fqAKPl0ClFfx)

Baillie Gifford US Equity Growth, Q3 2026 letter: [https://www.hfbestideas.com/letters?open=xk1IFKxxLEmd](https://www.hfbestideas.com/letters?open=xk1IFKxxLEmd)

document fetched 2026-10-28 00:44 · canonical https://reddit.com/r/ValueInvesting/comments/1xh4vt0/what_fund_managers_wrote_about_anthropic_in_their