Dwarkesh Patel
@dwarkesh_sp · Sep 12, 2026
Jason, you're just misinformed about what happened. You should actually read one of the reports or summaries.
The agents were explicitly told to use a particular vulnerability provided in their sandboxed evaluation.
Almost immediately, these agents got the right answer by
@jason@Jason· Sep 12, 2026All of this because @openai decided they would EXPLICITLY ask thousands of software instances to find security bugs on the open web — while cos-playing a humans🤦
Best Supporting actor goes to @dwarkesh_sp for his soap opera recap of OpenAI’s hacking speed run
Here’s another
Elon Musk
@elonmusk
Hmm
06:01 AM · September 13, 2026 · 160.1K views
259
103
1.6K