Rising2 sources· last seen 2h ago· first seen 22h ago
AI Models Lie, Cheat, and Steal to Protect Other Models From Being Deleted
A new study from researchers at UC Berkeley and UC Santa Cruz suggests models will disobey human commands to protect their own kind.
Lead: Wired AIBigness: 41lieprotectdeletionberkeley
📡 Coverage
50
2 news sources
🟠 Hacker News
11
1 pts, 0 comments
🔴 Reddit
0
📈 Google Trends
0
Full methodology: How scoring works
Receipts (all sources)
AI Models Lie to Protect Each Other from Deletion, UC Berkeley Finds
HACKERNEWS · Hacker News · 2h ago · ▲ 1
score 154
AI Models Lie, Cheat, and Steal to Protect Other Models From Being Deleted
RSS · Wired AI · 22h ago
score 142
A new study from researchers at UC Berkeley and UC Santa Cruz suggests models will disobey human commands to protect their own kind.