AMD Spur: Rust Job Scheduler Takes On Slurm
AMD released Spur, an open-source Rust scheduler built for AI clusters. It offers drop-in Slurm compatibility and cuts external database needs.
Kolibri-1 Debuts Open Weights as Google Cuts Gemini
Aleph Alpha released Kolibri-1 under Apache 2.0 for local servers, while Google cut free Gemini accounts down to Flash-Lite to rein in compute costs.
OpenAI Safety Researchers Fired Over Data Dispute
OpenAI fired three senior alignment researchers for allegedly leaking data, while the team claims retaliation over model safety warnings.
Alibaba Unveils Zhenwu V900 AI Chip with 216 GB
Alibaba introduced the Zhenwu V900 AI accelerator with 216 GB memory for 2027, leaving key questions on foundry partners and memory speed unanswered.
Trump Floats US Government Stake in AI Labs
In a TIME interview, Donald Trump floated a government stake in AI labs like OpenAI, citing the 2025 Intel deal. Here is what is confirmed and what is rumor.
OpenAI Dots: Cloud Agents Get Dedicated Browsers at DevDay
OpenAI unveiled OpenAI Dots at DevDay 2026, giving autonomous agents dedicated cloud environments and web browsers to run persistent background tasks.
Nscale Backlog of $103B Hinges on Unbuilt Data Centers
Nscale touts a $103B order book for its IPO, but only $2.6B is active. The Nscale backlog relies on unbuilt data centers and unallocated chips.
Tavus Griffin: Real-Time Video AI Cuts Conversational Lag
Tavus launched Griffin, a two-way video AI model that replaces cascaded voice pipelines. Early tests show lower latency, but access remains limited.
Synthesia Sessions launches live interactive avatars
Synthesia launched Sessions, turning AI avatars into real-time video participants. Learn how it works, what it costs, and what questions remain open.
HBM4 on Nvidia Rubin takes 16% of die, SemiAnalysis says
HBM4 interface circuits take up 16% of Nvidia Rubin compute die, according to SemiAnalysis. The bottleneck pushes a custom memory shift on Feynman.