Particle.news
Download on the App Store
18 ARTICLES
·
last mo.

Anthropic Safeguards Research Lead Resigns With Public Warning That ‘The World Is in Peril’

His exit sharpened questions about safety influence inside fast‑expanding AI labs.

Image
Getty / Futurism
The head of Anthropic’s Safeguards Research team said that “throughout my time here, I’ve repeatedly seen how hard it is to truly let our values govern our actions.”
A number of high-profile AI staffers have decided to call it quits, with some explicitly warning that the companies they worked for are moving too fast and downplaying the technology’s shortcomings.

Overview

  • Mrinank Sharma left Anthropic on February 9 and posted his resignation letter on X, which drew about 1 million views by Monday afternoon, according to Forbes.
  • He led the company’s Safeguards Research Team from 2023, working on AI sycophancy, defenses against AI‑assisted biothreats, and a recent study reporting thousands of daily chatbot interactions that may distort users’ perceptions.
  • In his letter, Sharma wrote that it is hard to let values govern actions within fast‑moving organizations and said he plans to pursue writing and a poetry degree.
  • Anthropic told CNN it was grateful for Sharma’s work and clarified he was not the head of overall safety nor responsible for broader company safeguards.
  • The departure comes as Anthropic rolls out Claude Opus 4.6 and Claude Cowork and is reportedly in funding talks that could value the firm at $350 billion, with other recent staff exits also noted in coverage.