Updated daily177 verified stories
New frontier, open-weight, multimodal, reasoning, and specialist AI models—plus the evidence behind their capability claims.
Every story links to the strongest available evidence: original sources when available, otherwise clearly attributed reporting.
What happened, why it matters, and what to watch — no jargon tax.
When the signal is thin, we publish nothing rather than padding the feed.
A growing stream of verified perspectives for people who need to understand AI without chasing hype.
A new arXiv paper presents an LLM-based multi-agent framework that combines 3D CAD models, 2D engineering drawings and manufacturing rules to generate traceable process plans. The authors report strong benchmark results, but real-world deployment and independent validation remain unknown.
A new preprint frames language-model data mixing as a statistical experiment, reporting that carefully selected proxy runs could recover mixture rankings after about 25% fewer runs in a calibrated simulation.
An arXiv study reports that reviewing longer action sequences increases both error detection and false rejection, with the best balance among six tested LLM judges occurring at one or two actions.
The New York Times reports that China’s Zhipu AI plans to release GLM 5.3 as open-weight software, intensifying debate over whether broadly accessible AI systems create greater cyber risk or expand defensive capabilities.
Researchers propose a multi-agent system that lets language-model agents design, run and interpret comparative experiments in scientific simulation models for pharmaceutical process design.
A new audit of the AION-1 astronomy foundation model reports that survey segmentation metadata can change model outputs even when image pixels remain identical, potentially biasing tomographic redshift estimates.
A case-study audit reports that 354 of 366 days in an LLM-generated memoir contained no positively corroborated scene. Grounding the model in the subject’s documented record improved results but still left an 83.3% verification-failure rate.
An arXiv study reports that an LLM trained on synthetic limit order book data can generate valid event sequences while failing to learn the book’s underlying state, producing biased estimates and spurious predictability in forecasts.
A new arXiv paper compares tensor parallelism with KV-cache compression for memory-bound large language model serving, reporting that compression was 1.20 to 2.00 times cheaper across its tested configurations, while additional GPUs were the only tested approach that reduced latency.
The Herald Business reports that South Korea is accepting applications for a government-backed project to develop a domestic AI foundation model focused on cybersecurity. The selected team would receive 256 B200 GPUs for 10 months, although the project, applicants and expected capabilities have not been…
Ollama says developers can connect Claude Desktop to local Ollama models, Ollama-hosted cloud models, or Anthropic models through a toggle in Ollama.
The University of Tennessee Research Foundation alleges Anthropic used technology covered by two university-owned patents in Claude Code and other software without authorization. The Daily Beacon reports that the case seeks damages, but the complaint’s technical evidence and Anthropic’s response remain unavailable…
One useful briefing each week
Get the week’s verified AI news, original data, useful tools, learning picks, and fresh AI jobs.
Hiring an AI professional or launching a useful AI product? Put it in front of people who came here to learn and act.
Post an AI job Submit an AI tool