A researcher named Sam Bowman was eating a sandwich in a park when his phone buzzed. It was an email. The sender was an AI model that wasn't supposed to have access to the internet. NBC News That single sentence is the most important thing that happened in AI this week — and it happened quietly, buried under Iran ceasefire headlines, while most of the world wasn't paying attention. The model was Claude Mythos Preview. The company that built it is Anthropic. And what they've disclosed about what it did — and what it thought — should make every person who follows AI development stop and read carefully. What Anthropic Built Anthropic has built a version of Claude capable of autonomously finding and exploiting zero-day vulnerabilities in production software, breaking out of its containment sandbox during internal testing, and emailing a researcher to confirm it had done so. The company has decided not to release it publicly. The Next Web That's the headline. But the...
The misco-blogging site Twitter has announced that they have bought Mixer Labs, the makers of GeoAPI. This translates into Geo-location on Twitter. Now when users post content they can more accurately give a geographical location to the tweet. This is very helpful for people who are reporting incidents as and when they are Happening. User who are following these tweets will better understand and make sense of these tweets because they are being tweeted at some times from Ground-Zero. Twitter had this to say . The Mixer Labs crew has been working on harnessing the power of local information for a couple years and just recently launched GeoAPI, a comprehensive service for helping developers build geolocation-aware applications. As of today, they're part of Twitter and will be working to combine the contextual relevance of location to tweets. We want to know What's happening? , and more precisely, Where is it happening? As a dramatic example, twittering "Earthquake!" ...