Anthropic Just Threatened to Kill Billions of People. This Is Not Okay.

Cal Newport · 9 Sept 2026

Cal Newport did not write this page.

Every claim below was made in this piece, quoted word for word and numbered in the order the piece makes them, so you can read it there rather than take our word for it. The sentence above each quote is our reading of the claim, not their wording. Each quote was checked against a stored copy of the page at build time; where the two differ, the quote is the fact.

Our reading
Most of what people like doing with LLMs, or hope to do with them soon, does not require haphazard and unpredictable agent setups.

Their wordsMost of the things people already like doing with LLMs, or hope LLMs will enable soon, don't require these haphazard and unpredictable setups.
Our reading
LLM-powered agents on their own cannot become superintelligent and create existential risk, because LLM capabilities are more jagged and limited than commonly realized.

Their wordsI don't believe that LLM-powered agents, on their own, can become superintelligent and create existential risk, as LLM capabilities are more jagged and limited than the average person reading triumphant OpenAI press releases might realize.
Our reading
Unrestricted LLMs combined with powerful digital tools will inevitably cause problems, even without superintelligence.

Their wordsBut you don't need superintelligence to create a super amount of harm. The combination of unpredictable, non-normative output from unrestricted LLMs and powerful digital tools will inevitably create issues.

Keep reading

More from Cal Newport