Full-text search across 26027 articles. Combine with topic and date filters; results sorted by relevance.
SpacebarBench, an October 2026 benchmark from Spacebar, evaluates six model configurations on fourteen consecutive user requests within a single persistent collaborative workspace, scoring delivered artifacts on a 100-po…
Notion said in an October 19 blog post that its agents can now run for more than 24 hours on a single task, up from roughly five minutes of autonomous work. Notion attributes the increase to moving every agent session on…
Trida AI published an analysis arguing that long-horizon AI agents fail not from context-window limits but from losing track of where work stands, and that a compact structured "map" of goal, current position, decisions,…
Humans&, the AI lab that raised $480 million from backers including Nvidia and Jeff Bezos before shipping a product, released its first model, Ampersand v0.1, trained from Z.ai's (Zhipu) open-weight GLM 5.3 rather than b…
A Linux sandbox for untrusted AI coding agents is a stack of partial mechanisms — cgroup v2, user and other namespaces, mount namespace with pivot_root, capabilities plus no_new_privs, Landlock and seccomp-bpf — that all…
Octave built a multi-tenant AI foundation on Amazon Bedrock using Cohere Embed v4 that embedded 2 billion documents and cut deployment time per use case by 50 percent, the company reported in an AWS blog post. The Relian…
MiniMax released M3.1 Flash on October 19, 2026, and independent benchmarking by Artificial Analysis reported by VentureBeat found the model's Intelligence Index score rose to 42 from 29 for its predecessor while average…
A single operator used AI agents to attempt intrusions against more than 1,200 websites and break into sites belonging to over 100 companies at an estimated AI cost of about $5,000, according to a Reuters report cited in…
A developer argues that production AI agent failures commonly blamed on the LLM are actually distributed systems failures, since an agent spans an orchestrator, LLM API, tool executors, vector stores and state stores, ea…
A shared RTK pipeline named laya-delegate cut mean cost per trial 13.9% across 48 SkillsBench trials on Claude Opus 5.5 run on 2026-10-18 and 2026-10-19, with reward unchanged at 0.81, according to self-reported numbers …
CrowdStrike in September launched SafeMind, a cybersecurity tool that uses custom AI models and harnesses to attack simulated copies of customer infrastructure while defensive AI systems detect, repel and patch the attac…
The clavity repository's latest Antigravity release ships a broken "--prompt-interactive" option, and its maintainer's documented workaround is to paste the contents of the agy-pairing-INSTALL.md file into Antigravity's …
The US Department of Homeland Security on Monday issued an order barring federal civilian agencies from using products from Chinese AI lab DeepSeek and requiring them to remove existing implementations, covering "all Dee…
Construct, an AI agent platform, published a guide showing how to run scheduled AI agent tasks in its cloud workspace so jobs fire with the laptop closed, with plans priced at $9 a month for 3 scheduled tasks, $59 for 10…
The Department of Homeland Security ordered federal civilian agencies to remove all DeepSeek-affiliated products, including third-party services built using DeepSeek, from their IT systems and barred future purchases of …
Induction AI's testing of Gemini 4 Argon found Google's new model caches prompts predictably at the documented 4096-token threshold, unlike Gemini 3.1 Pro Preview, which the company found dependable only after 9000 token…
Technical writer CT Smith told a podcast with Fabrizio Ferri-Benedetti that a layered pipeline of Vale hooks, skills, model reviews and sandbox fixtures lets Claude Code handle about 40% of her documentation tickets, a f…
Strot released an open-source Python package, strot-ai, that uses an LLM once at compile time to build a frozen local decision artifact, then classifies, scores and routes records on CPU at roughly 60 ms per record with …
DeepSeek opened its web search and page-fetch APIs to developers on 17 Oct 2026, pricing each at $0.75 per 1,000 successful requests, with failed requests, empty search results and timeouts unbilled. DeepSeek's own searc…
Emeraldleaf released okl (Observed Knowledge Ledger), a CLI installed via `pipx install 'observed-knowledge-ledger[mcp]'` that briefs AI coding agents on a repository's conventions, architecture decisions and checks befo…