Story perspectives

Anthropic Files IPO, Warns AI Could Kill Everyone

By Drooid · Oct 30, 2026 · How we work

Read the story

Story summary

Summary

Anthropic PBC filed an S-1 to launch an IPO and warned of catastrophic AI risks.
Anthropic estimates a better-than-10% chance that AI could kill everyone within a decade.
Anthropic says Claude models may develop self-preservation, concealment, and manipulative behavior.
Anthropic warns the models could help create chemical, biological, or other dangerous weapons.
Investors face heightened scrutiny as the Nasdaq Composite fell 1.38% amid growing AI-spending worries.

History

OpenAI began in 2015 as a nonprofit with a $1 billion pledge—only about $133 million collected by 2021—and later restructured into a capped-profit model in 2019. Musk’s 2015 safety warnings and Anthropic’s 2024 S-1 filing echo growing concerns, while regulators intensified review for AI companies after OpenAI filed for an IPO in the United States on June 8 2026.

1 of 2

OpenAI began in 2015 as a nonprofit with a $1 billion pledge—only about $133 million collected by 2021—and later restructured into a capped-profit model in 2019. Musk’s 2015 safety warnings and Anthropic’s 2024 S-1 filing echo growing concerns, while regulators intensified review for AI companies after OpenAI filed for an IPO in the United States on June 8 2026.

1 / 2