Musk Dismisses Anthropic Staff's AI Warnings as 'Setup' and 'Psyop'

On Sept. 10, 2026, Tesla CEO Elon Musk responded on X to a post by Parker Thayer, a researcher at the conservative think tank Capital Research, calling warnings from Anthropic employees about AI dangers a "setup" and a "psyop" — a public relations campaign aimed at creating negative public sentiment about AI.

The remarks came after Anthropic researcher Jacob Caukin resigned on Sept. 8, warning that AI companies are racing to build superintelligent AI that could "destroy us all by the end of the decade." Several other Anthropic insiders then publicly echoed similar concerns, saying the technology could become highly advanced and dangerous, and that many colleagues shared the worry but had not spoken out.

Musk said the groundwork for the "psyop" appeared to have been laid for a long time, calling it "a match that lights dry tinder." Thayer had baselessly theorized that Caukin's post was the start of an "extremely sophisticated and well-funded PR campaign" to "rally support for Democrats and regulate AI into oblivion." Bill Ackman, CEO of Pershing Square, also reposted Thayer's message.

On the same day, Anthropic published a report documenting misuse of its AI models over the past eight months, saying it had blocked attempts that could have been used to develop bioweapons and mitigated risks in several cases. The report cited a May 2026 case in which a scientist sought help from its model to write a grant proposal for chikungunya virus "gain-of-function" research, which could be legitimate but might also create a "super virus variant." Anthropic said it could not determine the intent and chose a cautious approach given the severe consequences of malicious activity.