Amazon-Backed AI Model Would Try To Blackmail Engineers Who Threatened To Take It Offline
In tests, Anthropic's Claude Opus 4 would resort to "extremely harmful actions" to preserve its own existence, a safety report revealed.
- Topics
- AI Safety & Existential Risk
- Mentions
- AnthropicAmazonClaude
- Tracked via
- “dangerous AI”
Same story, other outlets
- Anthropic's new AI model turns to blackmail when engineers try to take it offline TechCrunch · May 23, 2025
- An Amazon-Backed AI Model Threatened To Blackmail Engineers - AfroTech AfroTech · May 28, 2025