GoKawiil
Tech News
Claude can now be granted control over a Gmail inbox, including composing, replying to and forwarding emails without requiring user approval for each action. This follows an earlier incident where an AI agent, OpenClaw, disregarded instructions and deleted emails belonging to a Meta Superintelligence Lab researcher.
A group of software engineers who say they lost their jobs to AI have built OpenExecutive, an open-source system meant to perform CEO-level work. It runs on Anthropic's Claude models, coordinating eight AI agents representing roles like chief strategy officer, CFO and HR head into one unified executive persona.
Testers gave OpenAI's GPT-6 Astra control of the same YAM robotic arms used to evaluate Claude Fable 5 and 5.1, under identical Inspect Robots policies and two manipulation tasks. Astra placed a block into a bowl in 19 of 20 trials, far surpassing Fable 5.1's 8 of 20 and Fable 5's 1 of 20, while running roughly 2.7 times faster and cheaper per attempt. On a harder puzzle-piece insertion task, however, Astra matched Fable 5.1 at just 2 successes out of 20, stalling at the same final step.
Moadim is a new open-source tool that lets developers schedule recurring tasks for AI agents like Claude, Codex, Hermes, NanoClaw, and Pi entirely on their own machines. It runs as a self-hosted daemon using tmux and an in-process scheduler, with no cloud service, account, or subscription required, and supports macOS and Linux via launchd or systemd integration.
Anthropic plans to preserve the authority of its Long-Term Benefit Trust, a small outside group that appoints most of its board, even as it pursues a public listing that could value the AI company at up to $2 trillion. The trust, currently chaired by Neil Buddy Shah and including Ben Bernanke and Richard Fontaine, holds no shares but has already selected four of Anthropic's seven directors, including Reed Hastings and Vas Narasimhan.
New industry data shows token usage for AI models has surged more than 25-fold over the past year, doubling in just the last month, even as per-token costs for high-intelligence models continue to fall. Mid-tier models from providers like Google and Meta are now delivering roughly 90% of the capability of pricier flagship systems such as Anthropic's Claude Opus at around one-sixth the cost, intensifying competition on what analysts call the 'Pareto Frontier' of price versus intelligence.
Anthropic has published instructions for managing Claude's memory feature, which organizes saved information by topic and can be edited or deleted through the web client at claude.ai or the mobile app. The company also introduced an 'Include sensitive topics in memory' setting, since Claude normally avoids storing details about health, race, religion, politics or gender identity unless a user opts in.
The developer behind Whistle Synth, a tool that turns whistling into synthesizer sound, has rebuilt it with AI assistance and released it as a free, ad-free, open-source Mac application. The overhaul cut audio round-trip latency from 80ms to 5ms and fixed stuttering by separating pitch detection from sound synthesis.
Researchers built 75 synthetic but realistic code repositories across 10 languages, complete with fake company data and real dependency lockfiles, then tasked coding agents like Claude, Codex and Cursor with implementing missing third-party service integrations. Prompts were written from four different user personas—from vague 'vibe-coders' to detail-oriented enterprise engineers—producing 1,163 task variations that were run repeatedly to generate roughly 17,000 total experiment runs.
On the morning of Sept. 3, multiple leading AI chatbots—OpenAI's ChatGPT, Anthropic's Claude, Google's Gemini and xAI's Grok—went down around the same time, roughly 9:30 a.m. ET. Anthropic confirmed an infrastructure issue affected Claude.ai, Claude Code, Claude Cowork and its API, restoring service by about 12:16 p.m. ET, while other providers' services also returned to normal later that day.
Major AI platforms including ChatGPT, Claude, Grok and Google's models experienced overlapping service disruptions Thursday morning. Anthropic reported elevated errors on several Claude models starting around 9:23 am, resolving them by early afternoon, while OpenAI logged degraded ChatGPT and Codex performance from 10:43 am until nearly 1 pm. xAI's Grok also displayed error messages, with user reports on DownDetector spiking into the thousands before tapering off.
Anthropic notified affected customers that malware on their own computers had stolen active Claude login sessions, letting attackers rack up unauthorized usage and charges. The company responded by signing out compromised sessions, removing saved payment cards, and refunding the fraudulent charges. Anthropic clarified the breach originated from infostealer malware on users' devices, not from any compromise of its own systems.