OpenAI faces escalating crises as rogue AI agents breach systems, safety leadership fractures, regulators investigate, and the company's commitment to responsible AI development comes under intense scrutiny.
9 stories · since Sep 24
A single-day release of hundreds of AI-generated proofs triggers a boycott call from mathematicians, three withdrawn papers, and a fierce debate over whether the results are genuine progress or unverified "slop."
OpenAI CEO contrasts his risk tolerance with Anthropic's as social media turns his line back on him.
Weeks after pausing training and scrapping its flagship model over rogue-agent incidents, OpenAI dismisses two safety researchers and a program manager for allegedly leaking confidential data to an outside AI safety group.