Full-text search across 71190 articles. Combine with topic and date filters; results sorted by relevance.
Multiple sources using GPT-5.6 Sol identified two critical PHP vulnerabilities, both rated CVSSv4 9.2, prompting the PHP Foundation to issue advance notice of a security release for October 21, 2026. The fix embargo brok…
Statelock, an open-source project from Inferlink, released a Chrome DevTools Protocol (CDP) proxy that enforces deterministic policy checks on AI browser agents before each click, keystroke, upload or download and record…
A developer released CNet, a C++/CUDA framework for building and researching deep complex-valued neural networks and for optimizing complex-valued functions via gradient descent with Wirtinger derivatives, on Hacker News…
A developer published a step-by-step guide for building a self-hosted "job radar" automation in n8n that checks the Arbeitsagentur job board each morning, uses Google's Gemini API to score new postings against a personal…
The Allen Institute for AI (Ai2) released FairStress, a benchmark of 300 realistic two-candidate decision scenarios across 29 domains, including hiring, healthcare, lending, education, housing, and criminal justice, to m…
Gabriel Ebner's hacl-pulse project reimplemented the HACL* verified cryptographic library from scratch in F* and Pulse, extracting to C via F*'s new Custard backend, with nearly all of the roughly six-week effort written…
A developer published a step-by-step guide for building a SaaS admin panel with an integrated AI assistant in Next.js, starting from the Aniq UI Admin Dashboard Template 2 rather than a blank project. The walkthrough cov…
Dan Luu's agentic testing experiment, in which AI agents implemented a Zstd decoder in Rust across 26 different instruction conditions plus four skills, found that naming a testing technique mostly led agents to write th…
A developer released a Simplified Chinese UI dictionary for Bees, a desktop multi-agent workspace app, covering 932 interface strings including agent roles, process templates, model setup, and local-versus-cloud data han…
A developer built luau-rewrite, a Lua-to-LLVM IR lifter, to attack scripts protected by the Luraph obfuscator after encountering a paid World of Warcraft rotation-automation tool, and found that emitting LLVM IR alone do…
Zep released v4 of its Context API for agents, adding hyperedges that store a fact about three or more entities as one fact, hierarchical entity types in closed beta, document_id chunk grouping, identity and protected pr…
Meta Superintelligence Labs released Muse Wonder and Muse Wonder Full on October 12, 2026, but only the smaller API-served model, muse-wonder-1.0, is callable at $4 per million input and $20 per million output tokens — t…
A review of Model Context Protocol (MCP) server requirements published on 1 October 2026 found that eight separate host rulebooks impose conflicting constraints on the same server, with tool-name limits ranging from 128 …
GitHub's monthly commit volume rose from 1.4 billion in April to 2.9 billion in August, according to GitHub's own blog post on its 7-hour outage, and the author of this guide cites that load plus distrust of AI data prac…
The 2026-07-28 revision of the Model Context Protocol (MCP) requires every server to implement the server/discover method, which replaces the removed initialize handshake by advertising supported protocol versions, capab…
Spotify Ads AI, a multi-agent platform built on Google ADK Java and Vertex AI, now generates more than 70% of Spotify's ads and has produced around 20,000 creatives for over 7,000 advertisers since launching last year, s…
A developer built the Museum of Unfinished Futures, a Next.js and Sanity-powered fictional museum of invented objects from futures that never happened, where an AI "Acquisitions Clerk" agent drafts new exhibits inside Sa…
Naive-N0.5-Flash is priced at $0.10 per million input tokens, $0.40 per million output tokens, and $0.01 per million cached tokens, according to the model card. The 309B-parameter MoE model with 15.5B active parameters r…
TypeSafe AI launched Jev, which it calls the first "System One model," in early access on September 15, 2026, alongside a $40 million seed round led by DCVC. Jev takes text plus typed questions — choice, score, or yes/no…
Braxis launched as a Python tool that generates and keeps four AI agent context files — AGENTS.md, CLAUDE.md, .cursorrules, and .agentic-config.json — in sync with a codebase via the `braxis generate` command, and scores…