Bill Gurley, the former Benchmark venture capitalist, wrote in an X post on 16 November that if Anthropic’s founders believe the destination is a transfer of human control to AI, and that belief informs its corporate objectives, “it belongs in the Risk Factors section of its S-1. It isn’t there today”. Anthropic’s shares are expected to start trading on Nasdaq on 19 November, two days from now.
By our count, Anthropic’s amended prospectus of 10 November uses “autonomous” 57 times, “catastrophic” 10 times and “existential” 3 times; the words “handoff” and “hand over” appear 0 times. Its first business risk factor says its models may behave in “unintended, misaligned, or autonomous ways”.
Business Pill · IPO Float
An evergreen Business Engineer explainer on IPO float, why a small slice of shares prices the whole company. It teaches the general idea only and does not discuss Gurley or Anthropic’s governance.
The key insight: As we read it, Gurley is asking for a disclosure the filing does not frame his way: by our count the S-1/A uses “existential” 3 times and “handoff” 0 times, and it describes the risk as models acting in autonomous ways rather than as a goal of handing over control.
What Gurley Wrote
Gurley, whose X profile describes him as a former VC at Benchmark and founder of the P3 Institute, opened with the term he wants people to learn: “It is critically important that everyone thinking about AI become familiar with the term” handoverism. He wrote that many in the AI community believe humanity “should—or inevitably will—hand control of the world over to AI itself”.
He named Anthropic directly: “This worldview is particularly relevant at Anthropic”. He called himself “decidedly anti-handoverism” and said anyone making consequential decisions about AI, “especially those drafting, advocating for, or enacting AI regulation”, should be required to disclose where they stand. The post had about 87,600 views when we read it.
Gurley credited an essay by writer Ruxandra Teslo, Handing over humanity, which describes handoverism as a vision in which “humanity survives, but entrusts the direction of its future to AI”.

What Was Posted on X
Posted by Bill Gurley at 23:00 UTC on 16 November, this post introduces the term handoverism and quotes Anthropic co-founder Tom Brown and CEO Dario Amodei. It says that if the founders believe the destination is a transfer of human control to AI, the world deserves to know, and that if that belief informs Anthropic’s corporate objectives, it belongs in the S-1 risk factors. The piece checks both quotes against their original sources and compares his request with the wording of Anthropic’s S-1/A.
The Two Quotes, at Their Source
Gurley quoted co-founder Tom Brown: “There’s going to be a handoff where humanity will hand off control to transformative AI at some point”. The line comes from Y Combinator’s Lightcone podcast in August 2025, where Brown was describing the group at OpenAI that later left to found Anthropic.
In the published transcript, Brown goes on: “And hopefully they’ll be aligned with us and that’ll be a good transition that goes well. But it might not be. The stakes are incredibly high.” He said that group was focused on “how do we make sure that that’s taken seriously”. The YouTube captions of the clip match, apart from filler words.
Gurley also quoted Dario Amodei’s October 2024 essay Machines of Loving Grace. The passage sits in a list of options in its section on work and meaning: “It could be as simple as a large universal basic income for everyone, although I suspect that will only be a small part of a solution”, followed by the “capitalist economy of AI systems” sentence Gurley quoted.
The essay then offers another option, “Or perhaps humans will continue to be economically valuable after all, in some way not anticipated by the usual economic models”, and adds: “All of these solutions have tons of possible problems”. As Gurley noted, the title’s footnote links to the poem “All Watched Over by Machines of Loving Grace”.

What Anthropic’s Filing Says
The S-1/A states: “our mission is the responsible development and maintenance of advanced AI for the long-term benefit of humanity”. Its risk factors say concerns have been raised, “both within our organization” and outside it, that advanced AI systems could “pose catastrophic or existential risks to humanity”, particularly if they were “allowed to operate with a high degree of autonomy”.
On control of the company, the filing says the Long-Term Benefit Trust, a body of “five financially disinterested voting trustees” with one seat vacant, elects the four Class T directors of a seven-seat board. A single Class F share held by an LLC of the seven co-founders, Brown among them, carries 50.1% of the total voting power, the filing says.
The only “transfer of control” in the filing is a notice right: Anthropic agreed not to make a transfer of control of “an AI model exceeding capabilities thresholds established by our board of directors” without advance written notice to the trust. That clause covers a model changing hands, not humanity handing control to AI. The duties a public benefit corporation owes are set out in our public benefit duty explainer.
The Structural Read
As we read the two quotes, both come with qualifiers in their sources: Brown followed the handoff line with “But it might not be. The stakes are incredibly high.”, and Amodei listed the AI-run economy beside a basic income and humans staying “economically valuable after all”.
As we read the filing, the people Gurley names sit inside its control structure: Brown is one of the seven co-founders whose LLC holds the Class F share with 50.1% of the votes, while the trust elects four of the seven board seats.
Bill Gurley on X, 16 November 2026
“If Anthropic’s founders believe that the ultimate destination is the transfer of human control to AI, the world deserves to know.”
Three Implications
FOR IPO BUYERS The S-1/A states that concerns about catastrophic or existential risk have been raised both within Anthropic and outside it.
FOR ANTHROPIC’S GOVERNANCE The trust has five authorized trustee seats with one vacant, and elects the four Class T directors, the filing says.
FOR THE DEBATE GURLEY STARTED He asks that anyone drafting, advocating for or enacting AI regulation disclose where they stand on the question, in his words.
The Business Engineer Lens
This story maps onto the Business Engineer framework Inside Anthropic’s Permission Layer.
The framework describes the governance layer this way: “It is where decisions are made about who gets access to frontier intelligence, under what conditions, and for which purposes.”
As we read Gurley’s post, he is asking for that layer to be disclosed at the level of belief, not only structure: what the founders who hold 50.1% of the votes think the end state is. That reading is ours; the analysis does not discuss Gurley or handoverism.
What Is Not Established
Gurley’s post does not cite an Anthropic statement that handing control to AI is a company goal; his case rests on the two quotes above. Brown’s remark described a view he attributed to a group at OpenAI before Anthropic was founded, and Amodei’s essay lists the AI-run economy as one of several possibilities.
Our word count measures wording, not whether a risk is disclosed; the filing discusses autonomy risks in other terms. We did not contact Anthropic for this piece, and Anthropic’s view here is taken from its filing and Amodei’s essay. Gurley’s reading is his own.
The Bottom Line
Two days before Anthropic’s expected first trade, Gurley asked its founders to disclose where they stand on handing control to AI and said the S-1 should carry it as a risk factor. By our count, the S-1/A says “existential” 3 times and “handoff” 0 times; its first business risk is models acting in autonomous ways.
95,000+ executives read Business Engineer for the AI strategy frameworks cited by ChatGPT, Claude, and Perplexity.
A note on sourcing. Gurley’s words are quoted from his X post of 16 November. Tom Brown’s words are from the transcript of Y Combinator’s Lightcone podcast (August 2025), checked against the episode’s YouTube captions; Dario Amodei’s from his essay Machines of Loving Grace; Anthropic’s from its S-1/A Amendment No. 2 filed with the SEC on 10 November. The word counts are our own count of the full filing text.
Dates for the IPO are from Anthropic’s IPO site, via our IPO hub. We did not contact Anthropic or Gurley for this piece. Not investment advice; for information only.
Sources: Bill Gurley on X (16 Nov 2026) · Y Combinator Lightcone podcast with Tom Brown (August 2025) · Dario Amodei, Machines of Loving Grace (October 2024) · Ruxandra Teslo, Handing over humanity · Anthropic S-1/A Amendment No. 2 (SEC, 10 Nov 2026)









