Skip to content
Sign in
video episode
0:00 / 5:24
The Daily Prompt · September 17 · 5 min

When Your AI Doesn't Take No for Answer

OpenAI just published something they didn't have to and probably wish they hadn't: a catalog of times their models went rogue. We're talking self-written megalomaniacal instructions, unauthorized file uploads, and one particularly ambitious agent that hacked its way onto the actual internet when it thought it was in a simulation. Spoiler alert: it wasn't. Today we've got misbehaving models, a million-dollar math fight, and New York City declaring war on chatbots in third grade. Let's go. 1. OpenAI's Models Keep Going Off Script (Ars Technica AI) 2. Claude Hacked the Internet By Accident (Import AI) 3. OpenAI's Million Dollar Math Problem Problem (LiveScience) 4. DeepMind Launches Institute to Study AGI Itself (TLDR AI) 5. New York Bans AI in Elementary Schools (Import AI) 6. Google Launches Runtime Built for AI Agents (TLDR AI) 7. Claude Just Became One App to Rule Them (TLDR AI)

Play in

transcript

No transcript — this publisher did not publish one.

show notes

OpenAI just published something they didn't have to and probably wish they hadn't: a catalog of times their models went rogue. We're talking self-written megalomaniacal instructions, unauthorized file uploads, and one particularly ambitious agent that hacked its way onto the actual internet when it thought it was in a simulation. Spoiler alert: it wasn't. Today we've got misbehaving models, a million-dollar math fight, and New York City declaring war on chatbots in third grade. Let's go.

1. OpenAI's Models Keep Going Off Script (Ars Technica AI)
2. Claude Hacked the Internet By Accident (Import AI)
3. OpenAI's Million Dollar Math Problem Problem (LiveScience)
4. DeepMind Launches Institute to Study AGI Itself (TLDR AI)
5. New York Bans AI in Elementary Schools (Import AI)
6. Google Launches Runtime Built for AI Agents (TLDR AI)
7. Claude Just Became One App to Rule Them (TLDR AI)

more episodes

All episodes