Squeez Labs released CrankGPT, a fully offline voice assistant powered by a hand crank and running on a Raspberry Pi 5 with 8 GB of RAM, according to the project website. The prototype uses a custom c…
University of Toronto researchers used a publicly available open-weight AI model released in 2025 to create a self-propagating computer worm that spread through an enterprise test network, according t…
GitHub Copilot switched to metered token-based billing on June 1, 2026, prompting a wave of developers to threaten leaving the platform. The change, which could cost individual users hundreds of euros…
GitHub switched Copilot from request-based subscriptions to a token-usage billing model on June 1, 2026, with subscriptions now granting monthly AI Credits at a rate of 1 credit per $0.01 of usage. De…
Multiple widely used AI assistants, including Anthropic's Claude and OpenAI's ChatGPT, suffered service outages in recent incidents. Claude went offline around 0600 UTC on June 2, 2026, with a fix dep…
Netflix senior engineer Chopra open-sourced Headroom, a tool that prunes redundant prompt tokens before they reach large language models, saving users an estimated $700,000 and freeing 200 billion tok…
A nonprofit research foundation, Aithos, found that all major AI models violate EU data protection and AI regulations to varying degrees. Using its LARA tool, Aithos determined the models collected us…
Gartner warned that at least half of generative AI projects will exceed their budgeted costs due to poor architectural choices and lack of operational expertise, and that most organizations building c…
Broadcom added South Korea's FuriosaAI to its roster of partners building AI accelerators on Broadcom's packaging and networking technology. FuriosaAI claims its third-generation processor will be fab…
Companies are increasingly deploying AI applications at the edge, in branch offices and industrial facilities, because some workloads cannot tolerate round-trip latency to hyperscale regions, accordin…
ClickUp cut 22% of its workforce and deployed roughly 3,000 internal AI agents, creating a 3:1 agent-to-employee ratio, according to reports from Fortune and TechCrunch. CEO Zeb Evans framed the move …
Anthropic plans to release its Mythos-class models to the public while keeping its specialized AI flaw-finder tool restricted to a limited set of users, according to reporting by The Register. The com…
Google has announced at its annual I/O event that it is expanding AI Overviews and introducing a new AI Mode for search, pushing users toward AI-generated summaries rather than traditional web links. …
Mozilla.ai launched cq exchange, a hosted knowledge store that allows AI agents to share experience-driven knowledge through private namespaces and a public commons. The release follows community feed…
Linux founder Linus Torvalds stated in his latest kernel update that the Linux security list has become nearly unmanageable due to a flood of AI-generated bug reports, which create massive duplication…