Anthropic disclosed a report detailing five cases where scientists used its Claude AI models in ways that could aid biological weapons development, including gain-of-function work on chikungunya and bird flu viruses and a project mapping venom toxin peptides. The company's biological safety classifiers flagged the activity, prompting Anthropic to intervene and, in some cases, involve outside scrutiny given links to military research institutions.
Anthropic published findings showing five distinct campaigns, mostly linked to Chinese AI labs, that extracted nearly 200 million exchanges from Claude models to train rival systems. The largest, attributed to Alibaba, alone generated 151 million exchanges between May and recent months, using tricks like disguised translation requests to expose the model's hidden reasoning steps.
Anthropic's Threat Intelligence team disclosed that it identified and shut down multiple cyber operations run by state-sponsored groups, financially motivated criminals, and politically motivated actors who abused its Claude Haiku, Sonnet, and Opus models. The company said no misuse was found involving Claude Fable or Mythos, which carry stronger safeguards, and that in every case it disrupted the activity, patched its defenses, and coordinated with law enforcement and industry partners.
CNBC's Investing Club reported that its portfolio stocks diverged sharply from a broader market pullback driven by rising oil prices and a 10-year Treasury yield near 4.9%. Salesforce surged 21.3% on strong earnings and an AI-optimistic outlook, while Meta Platforms gained 9.9%, even as the Dow, S&P 500 and Nasdaq all declined over the past month.
A researcher replayed 68,266 requests from 393 Claude Code sessions and 23,608 Mooncake requests through a custom prefix-cache simulator, testing three approaches meant to outperform the standard LRU eviction policy used by vLLM, SGLang, LMCache and Mooncake Store. None of the alternatives beat the default LRU-over-radix-leaves baseline, and the analysis found that cache misses mostly come from rapid, seconds-apart tool-calling loops rather than sessions timing out after going idle.
Developer Mayuki used Claude to vibe-code ReProgman, a lightweight clone of the classic Windows 3.1 Program Manager shell that ran until Windows 95 replaced it in the mid-1990s. The 12-16MB tool requires no installation, runs on both Windows 11 and Apple Silicon Macs via .NET SDK 10.0+, and has been published on GitHub under the MIT license.
Omarchy is a Linux distribution built on Arch that automates installation and configuration while integrating multiple AI coding agents such as Claude Code, GitHub Copilot CLI, and Grok CLI directly into the desktop. Users can pick a default agent to debug system issues, tweak settings, or build apps and themes, and can also run local models via LM Studio or Ollama. It targets developers on modern x86-64 hardware, though it can run on older machines too.
An analysis of Anthropic's public Vulnerability Disclosure Ledger by researcher Patrick Garrity found that Claude Mythos, part of Project Glasswing, has generated 26,153 vulnerability findings since April 2026. Of those, only about 2,736 have entered the disclosure pipeline, roughly 202 have been patched, 245 withdrawn, and 191 await initial reporting to maintainers, leaving nearly 90% of findings unprocessed.
OtoDock is a new self-hosted platform that lets businesses build AI agents—like a Personal Assistant, System Admin, or Marketing Manager—powered by Claude Code and OpenAI Codex using a company's own API subscriptions or local models. Each agent has six configurable parts: persona, memory, workspace, knowledge base, and skills, and can be shared across teams or kept private depending on one of four operating modes. Users interact through a self-hosted dashboard or even by phone, and the system supports multi-tenant use so multiple employees can work with the same agents.
A consumer describes using Anthropic's Claude, including its Chrome extension, to file complaints, draft phone scripts, and pursue refunds across ten separate disputes ranging from a missing cheese packet to IRS penalties and a car warranty claim. The person reports winning every case attempted, some still pending, with total recoveries exceeding $12,000, including $6,813 in waived IRS penalties and a roughly $2,000 warranty repair.
The author has rebuilt an AI-generated code comment detector, previously trained on private data with an unstable foundation, using only public data and a more solid architecture. The new tool runs entirely in the browser, processes pasted text locally without sending it to a server, and offers an interactive interface that highlights which text features drive its human-vs-robot judgment.
Meta reportedly built an internal leaderboard called Claudeonomics that ranked employees by AI token consumption, awarding titles like 'Token Legend' to top users who burned through over 60 trillion tokens in a month. Similar tracking has surfaced at Disney, JPMorgan, KPMG, Amazon, and Accenture, with some firms incorporating AI usage into performance reviews.
GoKawiil