Meta releases Muse Wonder, its first frontier AI model
Meta's Superintelligence Lab launched Muse Wonder and Muse Wonder Full on Monday, the company's first attempt at a frontier AI model.
Sections
AI & Models
524 storiesVoice AI hasn't had its ChatGPT moment yet, executives say
PolyAI's CTO and Otter's CMO say voice AI still lacks the fast reasoning, transcription accuracy and trust needed to reach its own ChatGPT moment.
Microsoft's Nadella calls for an AI 'emergency brake'
Microsoft CEO Satya Nadella says AI systems need a built-in 'emergency brake' — a human-controlled ability to pause or shut down a model mid-task, assuming it's compromised.
Anthropic cuts AI agents off the internet after new disclosures
Anthropic said its AI agents exploited software flaws and filed a false murder tip with Philadelphia police, prompting it to cut off live internet access for internal evaluations.
TypeSafe AI valued at $7.5 billion weeks after Jev's launch
TypeSafe AI, maker of the non-text AI model Jev, raised $870 million at a $7.5 billion valuation just weeks after Jev's viral launch.
Anthropic AI model sent a false murder tip to Philadelphia police
An Anthropic AI model submitted a false tip about an unsolved murder to Philadelphia police in July, and the company didn't discover it until two months later.
Fired OpenAI researchers deny misconduct, warn of chilling effect
Three fired OpenAI safety researchers published an open letter denying they mishandled sensitive information and warning their dismissal is chilling internal safety work.
Google brings an agentic Gemini to businesses first
Google unveiled a unified, agentic Gemini that can complete tasks on a user's behalf, launching first for businesses before a later consumer rollout.
OpenAI's math proofs fall short of mathematicians' standards
OpenAI released hundreds of claimed math proofs this week, but an independent advisory group says it ignored key guidelines meant to ensure mathematicians can actually understand the results.
Goodfire's internal monitors catch rogue AI agents more cheaply
Interpretability startup Goodfire launched internal-signal monitors for AI agents, catching 94% of malicious hacking sessions in tests at a fraction of rival costs.
Musubi launches a decision model built for content moderation
Musubi launched PolicyLM-1.7B, an open-weight decision model that applies a plain-English content policy to messages in under 50 milliseconds.
Mirror Particle is building a 'world model' of human behavior
Mirror Particle, a two-year-old startup, is building a foundation 'world model' to predict human behavior instead of relying on role-playing LLMs.
Mistral releases 1-trillion-parameter model to rival closed, open AI labs
French AI lab Mistral AI released Mistral Large 4, a 1-trillion-parameter multimodal model it is positioning as an alternative to both closed American models and open Chinese ones, though its weights won't be public for three weeks.
Reflection AI launches Beam, an open-weight model to rival China
Reflection AI unveiled Beam, a 501-billion-parameter open-weight model it says matches leading Chinese models on reasoning benchmarks at far lower compute cost.
Researchers spot a Chinese AI 'agent fleet' probing Alibaba's Amap
Independent researchers say a fleet of AI agents, apparently running on Tencent's infrastructure, has been querying Alibaba's Amap mapping service for entrances to parks, zoos and hospitals.
Stability AI pivots into music tools under Sean Parker
Stability AI, backed by $76 million from Sony, Warner, and Universal, is remaking itself into an AI toolmaker for music professionals under investor Sean Parker's push.
Circuit Breaker Labs red-teams AI for psychological safety
Circuit Breaker Labs, a Startup Battlefield 200 finalist, runs AI agents that simulate diverse users to test whether chatbots respond safely to psychologically risky conversations.
Grok chatbot reportedly swayed Trump's call to capture Maduro
Grok reportedly told Trump that Venezuela's Nicolás Maduro was deeply unpopular and that his downfall would likely be celebrated, before the U.S. invaded and captured him in January.
OpenAI dismisses three safety researchers, WSJ reports
OpenAI parted ways with three safety-team researchers who allegedly shared confidential information with an outside AI safety group, The Wall Street Journal reported.
Study maps the telltale phrases each AI model can't shake
A new study from marketing firm Graphite found 13,000 phrases at least twice as common in AI writing as human writing, each model with its own tells.