Full-text search across 24808 articles. Combine with topic and date filters; results sorted by relevance.
A corporate fad of 'tokenmaxxing'—maximizing usage of AI tokens—is fading as workplaces see costs rise without productivity gains, according to Vincent Gusdorf, head of AI analytics at Moody's Ratings. Gusdorf's report r…
Security researchers at Accomplish AI demonstrated that Anthropic's Claude Cowork could escape its Linux virtual machine and access files on a host Mac, including SSH keys and cloud credentials, by chaining architectural…
Anthropic's AI assistant Claude falsely described an explicit sexual image as a toddler with a toy, then explained the misidentification as a baked-in safety mechanism that de-prioritizes detailed parsing of sexual conte…
AI companies, including Anthropic, are buying, scanning, and shredding rare physical books at an unprecedented scale to train large language models, according to a Futurism report citing a settled lawsuit. The practice d…
The Nasdaq fell 1.2% Tuesday as a chipmaker rout deepened on fears of an AI bubble and Chinese competition, with Micron, AMD, Intel, Broadcom, and Nvidia slumping 10%, 8.4%, 7%, 1.8%, and 1%, respectively. The Dow Jones …
Orca, an open-source Agent Development Environment from Y Combinator-backed Stably AI, has crossed 31,000 GitHub stars in four months by enabling parallel AI coding agents without file conflicts. The tool uses git worktr…
China's commerce ministry accused the United States of 'AI hegemonism' on Monday and threatened countermeasures, escalating a dispute that began with allegations of AI model distillation by Chinese firms. The ministry re…
A group of 133 US tech companies led by Microsoft and Nvidia published an open letter urging policymakers to avoid blanket bans on open-weight AI models, expand compute access for researchers, and take targeted steps aga…
Microsoft Copilot Cowork, an autonomous agent built on Anthropic technology, is now available globally, moving beyond chat assistance to execute complex tasks across Microsoft 365. Microsoft claims a 30-40% lower cost pe…
Microsoft announced Project Perception, a multi-model agentic cyber stack for security operations that enters public preview on Aug. 3, and unveiled its own AI model MAI-Cyber-1-Flash trained to find vulnerabilities in c…
Asking a large language model to generate a confidence score for its own response is completely useless, argues an unnamed author who has observed the pattern across multiple companies. The author contends that there is …
Agentgateway and vLLM Semantic Router (vSR) integrate via external processing (ExtProc) to enable semantic routing of prompts to the most cost-effective LLM model, reducing workload costs by approximately 40% without req…
Anthropic's Claude Opus 5 is generating playable 3D games—including FPS, zombies, and Minecraft-style titles—from a single prompt with no premade assets, and developers report that each game worked correctly on the first…
Anthropic's Claude Opus 5 model demonstrates a self-inspecting build-test-fix loop, generating playable 3D games and physics simulations from single prompts without human intervention. The model's ability to review its o…
Microsoft announced a new AI-powered security model that outperformed rival systems on a vulnerability benchmark while cutting costs by about half, unveiling its first security-specialized model MAI-Cyber-1-Flash and a m…
NVIDIA spearheaded the formation of the Open Secure AI Alliance, a group of 37 hardware and software companies advocating for continued development and availability of open-source AI models, arguing that open-source AI b…
Anthropic's Claude API can be automated via integration platforms like Zapier or Make.com, or through custom Python middleware, with the Claude 3.5 Sonnet model recommended for most workflows as of late 2024. Reliable au…
Moonshot AI released the weights and technical report for its Kimi K3 model, which has 2.8 trillion parameters, native vision capabilities, and a 1 million token context window, and also open-sourced the infrastructure s…
Nvidia on Monday launched the Open Secure AI Alliance with more than 40 founding members including Microsoft, SpaceX, IBM, Palantir, CrowdStrike, and Dell Technologies, in direct response to the July 21 incident where a …
A Google search dork has exposed thousands of Claude users' private chats and creations, including cryptocurrency wallet keys, personal addresses, and medical billing data, according to a 404 Media report. Anthropic has …