AIWadmin
Liquid AI closes its LFM2.5 line with a 24B on-device model
The open-weight release brings a 128K context window and doubled vocabulary to agents that run without a cloud…
Laplace builds AI red teams for robots and self-driving cars
An Ann Arbor startup wants to stress-test physical AI systems before they reach the real world, using AI…
Linus Torvalds ships Linux kernel patches partly written by AI
The Linux creator used AI assistance on a set of kernel patches, marking a shift in how the…
Blueprints AI says it can draft permit-ready plans in minutes
A Santa Monica startup trained on eight years of approved drawing sets is pitching AI that takes over…
Intel and Broadcom chiefs warn that capital may stall AI buildout
Both executives said financing, not chips or power, could become the binding constraint as returns take years to…
Spotify doubled its monthly code changes and rebuilt its guardrails
AI-assisted development pushed merged changes from 8,100 to 17,000 year over year, and the company is rewriting its…
Nvidia weighs a deal for open-weight lab Reflection AI
Talks could end in a purchase, a bigger stake or an acquihire, days after Reflection shipped its first…
Loops raises $3.5M to stitch retail AI into one layer
A16z's speedrun accelerator led a seed round for a startup that argues retailers should stop bolting on one…
Mistlabs raises $26M to let assistants talk to each other
Ambi left beta on Apple devices with backing led by Luminous Ventures and Shanghai's Huaqin Technology, ahead of…
Karpathy maps the widening gap in who actually sees AI
Anthropic's Andrej Karpathy says the public sees a chatbot while a few thousand insiders watch agents mint zero…
Sakana lets one model grade and improve its own work
A single language model that solves, judges and refines its own workflows posted real gains, but shared blind…
DeepMind proof system cracks open problems and clears peer review
AlphaProof Nexus solved nine Erdos problems and dozens of conjectures, then took the results through the journals OpenAI…
White House restores UK testers’ access to frontier models
The block on Britain's AI Security Institute was lifted after diplomatic pressure, days before the UK announced its…
Nvidia hands its Vera Rubin rack designs to the open compute project
Nvidia contributed Vera Rubin rack and tray designs to OCP while Cisco, Celestica and ASUS rolled out matching…
Microsoft bets its own silicon can carry Azure inference
Microsoft unveiled Maia 300 and said its homegrown accelerators will move to multi-gigawatt deployments beginning in 2027.