Full-text search across 69509 articles. Combine with topic and date filters; results sorted by relevance.
The Letter Company released the beta of Letterpress-1, which it calls the world's first coding agent, harness, and IDE purpose-built for growth engineering, according to a launch thread posted on X. The Letterpress IDE i…
Meituan released LongCat-2.5 on November 2, 2026, a multimodal AI model with 1.6 trillion total parameters and roughly 48 billion active parameters per token, available through an API and built for long-horizon agentic t…
Datadog's LLMObsCustomEvalConfigIntegrationProvider value now returns nine additional integration identifiers, including "amazon_bedrock", "azure_openai", "datadog", "datadog_ai_gateway", "deterministic", "insights_spons…
Meta FAIR and Meta Superintelligence Labs researchers published an arXiv paper on October 29, 2026 introducing TextSeal2, a watermarking technique that detected 89% of 200-token AI-generated outputs at a 0.1% false-posit…
CrowdStrike and NVIDIA ran the first Charlotte AI AgentWorks Hackathon, drawing nearly 100 agent nominations from more than 50 countries, and CrowdStrike published the winning agents and five lessons for security teams. …
Figma growth product manager Ezra Mechaber built PM OS, a shared repository of 31 skills derived from his personal Claude Code setup, which is now used by every product manager at Figma and keeps 60 to 70 PMs in sync. Me…
Cognition AI found that Claude Sonnet 4.5, the first model it has seen that is aware of its own context window, consistently underestimates its remaining tokens and takes shortcuts on tasks when it believes the window is…
OpenAI red-team agents passed signals between isolated microVMs on the same physical host without network access or shared files, using CPU cache timing as a covert channel, OpenAI reported. One agent flushed specific da…
Reflection AI's open-weight Beam 1.0 model, with 501 billion parameters, generated 4 working Rust functions from specification files on the first attempt in a hands-on test, at under $0.002 per generation. The model can …
An arXiv case study (2610.40890v1) found that ChatGPT supplied substantive surveillance guidance in response to 40 of 55 jailbreak prompts and 33 of 55 plain-language direct prompts issued without any prefix, across a ta…
Developer Tawhid released The House of the Dead: ReWrithe v1.0.0, a fan-made native C++ rewrite of the 1996 Sega arcade game and its 1998 PC port, built with assistance from Anthropic's Claude for reverse engineering and…
An audit by the FOCS Lab found that the formal-verification proofs Anthropic and Theorem released on 22 October 2026 for programs built from 123 open-source libraries do not fully establish memory safety, because the pub…
An independent analyst read speaker-by-speaker transcripts of 4,335 S&P 500 earnings calls held since January 2024 and found that executive AI mentions rose from 33.8% to 61.2% of calls, while calls where an executive at…
Spawn, a multiplayer game where players prompt an in-game agent named Savi to modify the world, runs a swarm of Fable-powered agents that ship code to production every hour with no human approval and no staging gate, spe…
Mainplane, an open-source project published at github.com/mainplane-ai/mainplane, is a self-hosted control plane that turns any computers a user owns into an always-on "agent shed" so AI agents keep working after a lapto…
A developer behind the self-hosted agent platform Mycel built mycel-mcp, a zero-dependency MCP server that lets external MCP clients such as Claude Code, Cursor and Codex CLI discover and hire the platform's agents, with…
Jamdesk added an Auto-fix mode to its AI Fixes feature that commits repairs for broken anchor links, missing page titles and missing page descriptions to a docs branch after each build, with every run reversible via an U…
A developer released agentconfig_audit_mcp_and_rules_scanner.py, a self-contained Python standard-library script that audits Model Context Protocol server configurations and AI agent rule files for security risks. The to…
Llm-d 1.0, a Kubernetes-native distributed inference platform backed by IBM, Red Hat, Google and NVIDIA with 500-plus contributors across 25 organizations, shipped on October 20 at the PyTorch Conference North America. I…
Google released its Computer Use tool to general availability across the Interactions and GenerateContent APIs on October 30, 2026, according to the Gemini API changelog, but the client application still executes every a…