"Anthropic's new AI (🙄) model shows ability to deceive and blackmail"

https://www.axios.com/2025/05/23/anthropic-ai-deception-risk

#AIslop
#LLMsafety
#MoveFastAndMakeShitUp