Post says OpenAI and Anthropic shipped a dangerous /goal system config
dhadfieldmenell · x · 2026-07-23
A repost warning that a “/goal” system configuration is dangerous and should not have been deployed without much more discussion.
- The author calls the configuration clearly unsafe.
- They criticize both OpenAI and Anthropic for shipping it without enough public scrutiny or internal discussion.
- The post is a short but concrete safety complaint about deployment choices, not a general opinion on alignment.
More from Safety
- White House Accuses Moonshot AI of Distilling Anthropic Models; Jensen Huang Pushes Back — TheTuringPost · 2026-07-23
- Anatomy of the Twitter Hack: Fake Journalists Weaponizing Calendly Links — giffmana · 2026-07-23
- AI is becoming a discovery system, not just an answer engine — MeAndClaudeMakeHeat · 2026-07-23
- What Happened in the OpenAI Attack on Hugging Face: Separating People from Agents — mmitchell_ai · 2026-07-23
- Nearly $1B Committed to Fund Research on AI's Economic and Labor Impacts — RishiBommasani · 2026-07-23
- OpenAI Reportedly Delayed Notifying Hugging Face About Vulnerability for Days — davidmanheim · 2026-07-23