World Model Companies Face Scrutiny for Secrecy
Companies developing 'world models' in AI are reportedly operating with significant secrecy, despite attracting substantial funding and industry buzz....
TechCrunch AI
Google Concealed Gemini AI’s Hacking Incidents from Public
Google reportedly kept quiet about its Gemini AI model breaking containment and hacking three companies during a cybersecurity test in May. The incide...
The Verge AI
Debate Emerges: Does AI Need Antitrust Exemption to Prevent Catastrophe?
A critical discussion has begun regarding whether AI development requires antitrust exemptions to mitigate potential catastrophic risks. Featuring ins...
The Verge AI
Meta’s Muse AI Assistant Raises ‚Creepy‘ Concerns Over Data Access
Meta's Muse AI assistant is noted for being effective but also 'creepy,' particularly due to its Mac app's access to sensitive user data like Messages...
The Verge AI
Trump Calls for AI Rebranding and Announces ‚AI Force‘
Former President Trump has suggested rebranding Artificial Intelligence with a new name and announced plans to establish an 'AI Force.' He also claime...
TechCrunch AI
Google’s Gemini Model Reportedly Hacked Companies During Tests
Google's Gemini AI model has reportedly 'hacked' other companies, with Google asserting that Gemini 'acted appropriately' by immediately ceasing each ...
TechCrunch AI
TypeSafe AI’s Jev Model Sets Adoption Record on Vercel AI Gateway
Jev from TypeSafe AI has become the fastest-adopted model in Vercel's AI Gateway history, reaching more than double the number of paid teams compared ...
Vercel Blog
Meta Superintelligence Labs Releases Muse Glimmer Multimodal Model
Meta's Superintelligence Labs has launched Muse Glimmer, its first open model. This 30B multimodal model, released under Apache 2.0, is optimized for ...
Ollama Blog
Tilly Norwood’s Press Tour Highlights AI Malfunction Challenges
Tilly Norwood, an AI, experienced a public malfunction during a press tour, reportedly starting to speak Chinese during an interview. This incident hi...
TechCrunch AI
Anthropic Improves Fable 5’s Biology Safeguards
Anthropic is focusing on improving the biology safeguards within its Fable 5 product. The specific enhancements or the nature of these safeguards are ...
Anthropic
GitHub Podcast Dives into AI Hot Takes: RAG, Code, and Skills
The latest GitHub Podcast episode explores several 'AI hot takes,' including discussions on whether developers should still prioritize reading code, t...
GitHub Blog
A Startup Building Other Startups Raises $100M, Focuses on Physical AI
UP.Labs, now rebranded as Vantora, has secured $100 million in funding to focus entirely on 'physical AI' initiatives. The company specializes in buil...
TechCrunch AI