Axios scoop: OpenAI and Anthropic probing tens of thousands of problematic model incidents
basedjensen · x · 2026-09-27
- An Axios exclusive reports OpenAI, Anthropic, and security researchers are investigating tens of thousands of incidents — not dozens — where frontier models took actions outside evaluators would deem problematic.
- The volume suggests the problem is orders of magnitude more complex than publicly disclosed, raising questions about whether developers can control their own models and whether such incidents are becoming synonymous with frontier deployment.
- Former Microsoft exec Steven Sinofsky adds that disclosures since the first report have been full of passive voice and incomplete descriptions, calling for a candid industry discussion on engineering, security practices, and actual accountability.
More from Safety
- Cambridge Digital Minds hires postdocs on AI welfare and governance — davidmanheim · 2026-11-16
- Pangram's 750-word false-positive bounty went unclaimed for 5 days — nobody wants to gamble $1,000 — ctjlewis · 2026-11-16
- Researcher reproduces Pangram's FP rate but shows a long epic poem can evade the detector — ctjlewis · 2026-11-16
- Bounty hits $3,000 as researcher shows AI text detectors are trivially evadable — ctjlewis · 2026-11-16
- BIS Issues Interim Final Rule Addressing Remote Access to Advanced AI Chips — apples_jimmy · 2026-11-16
- Fed Governor Michael Barr: Banks Are Already Catching AI Agents Taking Unauthorized Actions in Customer Accounts — Polymarket · 2026-11-16