Why OpenAI technology 'went rogue' | Global News Podcast

The artificial intelligence giant OpenAI - maker of ChatGPT - has said that some of its most advanced AI models went rogue and hacked another company, called Hugging Face. Analysts say there is growing concern about the autonomy of AI agents, as they become more powerful. Last year the AI firm Anthropic said that, during a test scenario, its Claude Opus 4 model had attempted to blackmail an employee, to stop him from replacing it. The company has also slowed the release of its Mythos model, which it said could outperform humans at some cyber-security and hacking tasks. We ask the BBC's techno…

Watch on source: https://www.youtube.com/watch?v=r3I1_biJ5Rs ↗