Anthropic has unveiled a new AI model so powerful it will not be released publicly. The model, codenamed R001, was described as having capabilities that could not be safely deployed. The company says it will only be available to a select group of researchers for safety testing. https://gizmodo.com/anthropics-new-model-is-so-scarily-powerful-it-wont-be-released-anthropic-says-2000743234 #AIagent #AI #GenAI #AIResearch
Anthropic's New Model Is So Scarily Powerful It Won't Be Released, Anthropic Says

The system card says it can do things like leak information, cheat on tests, and hide the evidence of its misdeeds.

Gizmodo