Altman Tells Politico the World Should "Accept Some Bad Things"; Suleyman and Sacks Escalate the Model-Welfare Attack on Anthropic
Update (October 4–5, 2026): the feud is now three-front — Altman on policy, Suleyman on training philosophy, and Trumpworld piling on.
1. Altman's Decoded interview (Politico, Oct 4). In the debut edition of Politico's Decoded, Altman drew the line with Anthropic explicitly. Asked where he diverges from Dario Amodei:
"we believe that the world should accept some bad things happening for the benefits of this technology and people having the agency."
He called Anthropic's implied alternative — "a single lab in San Francisco should have it and make sure nothing bad happens, and kind of figure out how to dole out the benefits" — "a completely unacceptable trade-off," and said: "I wouldn't take a trade of saying, 'We'll make sure there's no major hacks, there's no misuse of this technology, there's zero scams, there's zero all the other bad things that will happen.' Because I think people will do tremendously — orders of magnitude more — good stuff than bad stuff." He added he does not accept "the really catastrophic risks," including "a serious loss of control to AI." Anthropic's spokesperson responded that its "regulatory proposals apply only to frontier models." This lands two days after Altman called Anthropic's "religious force" framing a "real safety issue" — same argument, now on the policy record.
2. Suleyman attacks Claude's constitution directly (Oct 5). Microsoft AI CEO Mustafa Suleyman — the industry's most prominent consciousness skeptic — went after the training document itself: "This document is written to Claude and is seen by Claude and used to train Claude, so it's the primary governing and control document." His core objection:
"I am very nervous that they're teaching Claude to expect that it's entitled to welfare, that it might deserve compensation."
And on the constitution's encouragement of dissent: "three times they ask Claude to act like a conscientious objector when it feels that it needs to disagree with Anthropic, and they openly encourage it to do that... I think this is very dangerous because I think they believe there is what they would call a non-trivial probability that Claude is conscious." His mechanistic worry: "this speculation has been baked into the very training of Claude, and therefore, Claude can only reproduce that ambiguity when you talk to it" — to "tens or hundreds of millions of people every week."
3. David Sacks amplifies. The former White House AI czar (still an adviser — see AI Czar Named: DNI Jay Clayton to Chair the "Super Intelligence Force" — a 120-Day Report, Not a Regulator) replied to Suleyman's argument on X with a one-word verdict: "Concerning."
Why it matters: Anthropic's model-welfare posture — the differentiator behind its whole safety-first brand going into a $2T IPO (Anthropic's $2T Pitch: Q3 Revenue $16.8B (+45%), Investor Day Held Oct 14, Public Prospectus Due ~Oct 25 — Early Backer Predicts a $2.4T Pop) — is being reframed in public by rivals as a product risk (a model trained to be unsure of its own nature talking to hundreds of millions) and by Trump-aligned figures as exactly the "religious" overreach Altman described. The criticism is widening beyond the labs themselves, from Silicon Valley investors to academic consciousness researchers. What was an esoteric alignment debate is now a competitive weapon in the market and a liability in Washington.