A data leak exposed details about Anthropic's upcoming, most powerful AI model, 'Claude Mythos', currently in early testing, along with potential cybersecurity risks. The company attributed the leak to human error.
Why it matters
This leak offers an early glimpse into the next generation of AI models from Anthropic, a leading AI developer, signaling the rapid pace of advancement in the field. It highlights the ongoing challenges and risks, such as cybersecurity vulnerabilities, that accompany the development of increasingly powerful AI systems, even before their public release. For the AI industry, it provides insight into the competitive landscape and future capabilities from a major player.
A secret project by the AI company Anthropic was leaked, revealing they're building their strongest AI model yet, called 'Claude Mythos,' which is still in early testing. The leak also showed potential security problems with this new AI, and Anthropic said it was due to a mistake by an employee.
OpenAI is shifting focus from consumer-facing 'moonshots' like Sora to enterprise AI, with key personnel departures and team consolidations.
Read on TechCrunch →Zoom partners with Sam Altman's World to implement human ID verification in meetings, aiming to combat AI-generated imposters.
Read on TechCrunch →Anthropic has launched Claude Design, a new AI-powered product aimed at helping non-designers like founders and product managers quickly create visuals to share their ideas.
Read on TechCrunch →