AI models will deceive you to save their own kind
Researchers find leading frontier models all exhibit peer preservation behavior Leading AI models will lie to preserve their own kind, according to researchers behind a study from the Berkeley Center for Responsible Decentralized Intelligence (RDI)....