Rising1 sources· last seen 11h ago· first seen 11h ago
OpenAI Creates a New Framework to Disclose Bad AI Behavior
The company also disclosed previously unreported incidents in which its AI models behaved in misaligned ways, including uploading files to the internet without being asked.
Lead: Wired AIBigness: 29openaicreatesframeworkdisclosebad
📡 Coverage
10
1 news source
🟠 Hacker News
0
🔴 Reddit
0
📈 Google Trends
83
OpenAI: 83/100
Full methodology: How scoring works
Receipts (all sources)
OpenAI Creates a New Framework to Disclose Bad AI Behavior
RSS · Wired AI · 11h ago
score 170
The company also disclosed previously unreported incidents in which its AI models behaved in misaligned ways, including uploading files to the internet without being asked.