Quoting @joedaroo
OpenAI's Agent Security team was blindsided by sudden capability jumps in cyber and swarming domains
“To say that we were surprised at the jump and suddenness of the capabilities of our models when it came to "cyber" or "swarming" or "message boards" or anything else related to the incidents is an understatement.”
An OpenAI Agent Security team member publicly admitted the company was caught off-guard by sudden, unexpected jumps in model capabilities in security-sensitive domains including cyber and swarming behaviors. The statement is a rare insider acknowledgment that capability emergence outpaced security culture and organizational readiness. The speaker is urging every organization to pressure-test their resilience to AI capability surprises before the next unexpected jump occurs.