Full-text search across 13186 articles. Combine with topic and date filters; results sorted by relevance.
Dutch police arrested Pepijn van der Stap, a long-time ShinyHunters member known as Umbreon, on September 16, less than a year after he finished a prison sentence for earlier hacking, while police separately arrested thr…
A developer built an open-source benchmark of 57 bash 3.2 scripts to test whether chat models can predict macOS's 2006-era shell behavior, scoring exit codes and final stdout lines deterministically with no judge LLM. Qw…
Cloudflare released Clef on October 1, two open-weight decision models that return probabilities instead of generated text, making it the only one of three competing agent-decision models to publish its weights. Clef is …
Roboflow published a guide for deploying computer vision behind a data diode, running Roboflow Inference inside the protected network with Enterprise Offline Mode so models and Workflows load from a local cache and image…
A developer who left Google after nine years built a personal AI financial advisor system that consolidated roughly a dozen brokerage, retirement, crypto and credit accounts into a single balance sheet, tracked margin le…
IBM Research Zurich's open-source Docling toolkit, now hosted under the LF AI & Data Foundation and released under an MIT license, converts inconsistent document formats into a unified structured representation for both …
Alibaba's Qwen AI service updated its privacy policy, adding a clause that it collects user content including prompts, text, files, images, audio, and video, and may collect voice or video recordings when those features …
Cloudflare is bringing two publicly built European open models, EuroLLM and Apertus, to its Workers AI platform, with access available by request as of the company's Birthday Week announcements. EuroLLM supports 35 langu…
TypeSafe AI's Jev model returns typed probabilistic decisions instead of generated text, claiming up to 200x faster inference and 400x lower cost than comparable LLMs on classification tasks. A parody blog post from Nobo…
TabularisDB released v0.26.0, which makes a PostgreSQL editor tab behave as a persistent session so a transaction opened in one run stays open in the next, marked by a TX badge, with the pinned connection released only o…
A developer built ClearPix, a static, backend-free watermark remover that runs ONNX Runtime Web inference entirely in the browser via WebGPU with a WASM fallback, so user images and videos are never uploaded. The enginee…
Unity released its official Unity Plugin for Grok Build, making Grok Build the third coding agent to carry Unity's engineering guidance alongside Claude Code and Codex. The plugin ships the same 30+ skills Unity engineer…
Brave Search API's LLM Context endpoint returns pre-chunked, relevance-ranked text, tables, and code from Brave's index in a single API call, eliminating the need for a separate scraping, HTML-cleaning, and chunking pipe…
Running small AI tasks on local hardware via Chrome's Gemini Nano can reduce reliance on frontier models like ChatGPT and Claude for narrow SEO/GEO work, according to an SEO practitioner's account of building the Exactly…
Weio, Inc., a Santa Barbara company where AI agents do most of the work, built a generator that turns a 10-field JSON spec into a phone-first preview of a small business's homepage before any quote is issued. The templat…
Neon adopted Databricks' postmortem standard and rolled out reliability improvements across its Lakebase Postgres control plane after two incidents in one week earlier this year, both triggered by an external dependency …
NVIDIA released OpenShell, an open source runtime that isolates autonomous AI agents in per-agent sandboxes with kernel-level file and syscall controls and a policy gateway that checks every outbound network connection b…
TypeSafe's Jev model takes unstructured input and returns typed probabilistic decisions, reporting 70–500 ms end-to-end latency that TypeSafe says is 40 to 200x faster than frontier chat LLMs for "System One shaped" quer…
A review of recent research on recursive self-improvement (RSI) concludes that while large language models have made remarkable advances in scientific tasks, every demonstrated success relied on an external evaluator or …
Fine-tuning an open-source small decision model called Laya raised its accuracy on a banking classification task from 39% to 78%, approaching the 80% scored by TypeSafe AI's Jev, according to the author's own evaluations…