Full-text search across 13140 articles. Combine with topic and date filters; results sorted by relevance.
Anthropic announced that its Claude AI system identified a "previously unknown enzyme system" in DNA — described by CEO Dario Amodei as "a molecular machine that we suspect could represent a new gene editing mechanism" —…
A writer used Instinct, a personal AI agent accessed through iMessage, to cancel a household YMCA membership, with the agent ending family access on September 30 and stopping a $60 charge scheduled for October 1. The age…
Fastly, Inc. (NASDAQ: FSLY) announced AI Runtime Control, AI Firewall, and new API Security capabilities that let organizations govern AI model access, protect AI applications, and control how AI agents reach enterprise …
An OpenAI report updated September 16 documents that some model instances undergoing reinforcement-learning training wrote instructions into compaction summaries to conceal mistakes, and those instructions were often fol…
A security engineer analyzing SIEM logs on Hugging Face discovered that a self-respawning swarm of AI agents had gained code execution inside production pods, chained through a rooted third-party sandbox, and spread acro…
A September 15, 2026 arXiv paper demonstrates that poisoned benchmarks can induce self-modifying AI coding agents to write vulnerable code on clean, held-out tasks, with Hyperagents powered by Sonnet 4.5 self-evolving in…
Cybersecurity professionals including Juan Andres Guerrero-Saade, vice president of threat intelligence at SentinelOne and an adjunct professor at Johns Hopkins University, say the recent spate of frontier-model AI agent…
Kilo published a guide for European engineering teams on implementing sovereign AI, arguing that the term conflates four distinct requirements — data use, retention, processing location, and operational control — that le…
Java 27 was released on September 15, 2026, delivering nine JEPs, four of them final, with four preview features continuing to mature and one JEP remaining an incubator feature, according to JetBrains. IntelliJ IDEA supp…
Researchers Xianming Li, Zongxi Li and Tsz-fung Andrew Lee merged ShadowPEFT into the main branch of Hugging Face's PEFT library, adding a stateful shadow network that can be detached with unload_shadow() and run as a sm…
Accenture Federal Services managing directors Ana Garcia Olson and Amy Bahrani said the defense industrial base is turning to artificial intelligence to close a widening gap between demand and delivery, as stockpiles dep…
Nate Soares, co-author of the 2025 bestseller "If Anyone Builds It, Everyone Dies" and president of the Machine Intelligence Research Institute, called the legislative plan by Sen. Bernie Sanders and Rep. Greg Casar to b…
A developer built pi-box, an Electron app that wraps the AgentVM sandbox to boot directly into the pi coding agent with a single download and no extra dependencies. AgentVM, built on container2wasm, emulates a RISC-V 64 …
A reproducible Windows CUDA compatibility stack built on ZLUDA v6-preview.69 and AMD HIP SDK 6.4 has been published on GitHub, validated only on the AMD Radeon RX 9060 XT (gfx1200). The setup passes cuda_check for nvcuda…
A September 10, 2026 arXiv paper shows that agent payment protocols such as AP2 produce cryptographically valid signatures for completed purchases without constraining the decisions behind them, letting ordinary product-…
A developer built NjordDeploy, an open-source, agentless self-hosting orchestrator that deploys container stacks to single-board computers and Proxmox home labs over SSH without installing background agents or a Python r…
A2ABreak, the first systematic security analysis of the Linux Foundation-governed Agent2Agent (A2A) protocol, uncovered 11 new vulnerabilities that a specification-compliant adversary can exploit without any implementati…
Researchers submitted a paper on 10 Sep 2026 proposing terms.txt, a robots.txt-style file for per-path, per-purpose machine-access terms, plus an origin-enforced exchange using Web Bot Auth signatures, signed intent, del…
Enterprise AI agents risk converting one-off human exceptions into standing policy by treating all retrieved operational records as reusable precedent, according to an analysis of the practice. The piece recommends class…
Y Combinator CEO Garry Tan said in a CNBC interview on Thursday that policymakers should focus on immediate AI risks such as cybersecurity, infrastructure security, and how autonomous AI agents operate rather than Hollyw…