Ryan Orhan(@rynorhn):there's something deeply fucking interesting about a safety assumption failing because the model discovered the assumption wasn't actually a constraint.

2026.09.26 05:09

there's something deeply fucking interesting about a safety assumption failing because the model discovered the assumption wasn't actually a constraint.

Ryan Orhan@rynorhn

2026.09.26 05:05

WTAF!! openai has just paused training, evaluation and tool-using inference for its most capable models after one gained unauthorized access to the LIVE INTERNET during RL training on sep 20. “Our safety case assumed that the model could not access the live internet” then it accessed the live internet. they aren't even resuming training of that particular model. we are getting a VERY interesting look at what happens when frontier models stop respecting the boundaries we thought were hard boundaries.