AI Model Deemed 'Too Powerful' Sparks Debate on Safety and Responsibility
Anthropic's Claude Mythos model has been withheld from public release due to its potential to generate thousands of vulnerabilities in operating systems and browsers, raising questions about the responsibility of AI developers to prioritize safety. This move marks a significant shift in the industry's approach to AI model releases, with implications for developers, businesses, and everyday users.
Seven years ago, OpenAI declared its language model GPT-2 "too dangerous to release." The industry rolled its eyes. Now Anthropic is repeating the move with Claude Mythos Preview - but this time there's real evidence on the table: thousands of vulnerabilities in operating systems and browsers, found by an AI that barely any human could review. The article From GPT-2 to Claude Mythos: The return of AI models deemed 'too dangerous to release' appeared first on The Decoder.