OpenAI builds Control Room kill switch for its research models
AI SECURITYOpenAIredteam/daily · 1 source
- Severity
- Info
- Sources
- 1
What happened
OpenAI has built a central emergency stop system, called Control Room, that can halt its AI models in its research labs when safety or security problems appear. The company says it can stop a model from producing output about 35 seconds after someone decides to pull the plug, and it is working to make that instant.
Like I’m new · Written by AI from 1 source · Wrong?
What to do
- No action needed.
Written by AI from the sources, checked against them
Full story timeline · 1 update
Who reported · 1 outlet
The Register (Security)
Who and what is in it
Companies