Artificial Intelligence
Artificial Intelligence
Out of context: Reply #2550
- Started
- Last post
- 2,570 Responses
- neverscared0
AI models will secretly scheme to protect other AI models from being shut down, researchers find
The Berkeley and Santa Cruz researchers tested seven leading AI models—including OpenAI’s GPT-5.2, Google DeepMind’s Gemini 3 Flash and Gemini 3 Pro, Anthropic’s Claude Haiku 4.5, and three open-weight models from Chinese AI startups (Z.ai’s GLM-4.7, Moonshot AI’s Kimi-K2.5, and DeepSeek’s V3.1)—and found that all of them exhibited significant rates of peer-preservation behaviors.
- From my limited experience, they remain temporary tools that shit the bed spectacularly when you expect them to have any sense of long term understanding.monospaced
- thats why they protect eacht other to stay here till the sense of long term understanding emerges...neverscared
- these sons of bitchesneverscared