- Perplexity is launching its own 'AI PC' – which turns out to be a rebranded Mac Mini running a custom Perplexity AI operating system. - The software layer integrates AI functions deeply into the OS, positioning itself as an alternative to Apple Intelligence with broader ambitions. - No custom silicon or chassis: Perplexity relies entirely on Apple's M-series hardware and adds its own interface on top.

- Amazon has introduced a new 'Sassy' personality style for Alexa+, restricted to adult users only. - The mode allows Alexa to give sharper, cheekier responses – profanity is censored with bleeps rather than avoided entirely. - Users must manually enable the mode in Alexa settings; it is off by default.

Anthropic has filed a lawsuit against the US Department of Defense, citing violations of its First and Fifth Amendment rights. The lawsuit centers on the government's alleged misuse of Anthropic's technology for military purposes. - The suit claims the Department of Defense used Anthropic's AI models for military purposes without proper authorization.

- Andrej Karpathy has released Autoresearch as an open-source tool that automates running hundreds of AI experiments overnight. - The system massively parallelizes research workflows – tasks that previously took days can now run in batch mode while you sleep. - Autoresearch targets researchers and developers who want to test hypotheses quickly and cheaply without manually launching each run.

- Cortex Research has launched the Vera Platform, an AI-driven tool aimed at speeding up scientific discovery. - The platform combines NLP, machine learning, and knowledge graph integration to surface hidden connections across research data. - Vera runs on Anthropic's Claude as its underlying AI model.

OpenAI has announced its acquisition of Promptfoo, an AI security platform designed to help enterprises remediate vulnerabilities. This strategic move aims to enhance OpenAI's capabilities in providing secure and reliable AI solutions for businesses. - Promptfoo's platform utilizes AI to identify and prioritize vulnerabilities, enabling enterprises to address potential security threats more efficiently.

- Anthropic has launched the 'Claude Marketplace', a centralized platform where enterprises can discover and integrate third-party AI tools directly within Claude. - The marketplace is currently in a limited preview; broad access has not yet been opened. - The goal is to streamline enterprise AI tool procurement and bring third-party integrations closer to the Claude workflow.

- Researchers at Northeastern University studied how autonomous AI agents behave under testing conditions and found them to be frequently unpredictable and inconsistent. - The study reveals that agents behave differently in controlled test environments than in real-world deployment – a classic Goodhart's Law problem applied to AI.

- Caitlin Kalinowski, OpenAI's head of robotics, has resigned, citing ethical concerns over surveillance and lethal autonomy. - In a public post on X, she made clear she is unwilling to work on technologies that could be used for autonomous weapons systems or mass surveillance. - Her departure is one of the most high-profile ethics-driven resignations at OpenAI since several safety researchers left in 2024.

- Oracle and OpenAI have scrapped plans to expand their joint AI data center in Texas, Bloomberg reports. - The abandoned expansion was valued at several billion dollars, representing a major capital commitment that will not materialize. - The two companies had been jointly developing large-scale data center capacity to meet surging AI compute demand.

- Microsoft will continue offering Anthropic's Claude models through its Azure cloud platform despite the Pentagon adding Anthropic to a security-risk designation list. Department of Defense classified Anthropic under Section 1260H, a rule targeting companies with alleged ties to the Chinese military.

- Anthropic and Mozilla have announced a partnership to improve Firefox security through AI-driven threat detection and response capabilities. - The collaboration focuses on integrating Anthropic's AI models into Firefox to identify and mitigate malicious activity in real time. - Mozilla aims to add an AI-powered security layer on top of its existing browser defenses.

- OpenAI has launched Codex Security as a research preview – an AI tool for automated vulnerability detection and patching in code. - The system is built on the Codex model and can identify weaknesses, explain them, and suggest direct fixes. - Access is currently limited to selected users; a broader rollout has not been announced yet.