Anthropic has triggered alarm bells by touting the terrifying capabilities of “Claude Mythos”– with executives warning the AI model is so dangerous that it would cause a wave of catastrophic hacks and terror attacks if it was released to the wider public.
