Nvidia’s MediaTek Billions: Keeping AI Chips in Its Network
Nvidia is investing $3.5 billion in MediaTek. The deal shows how it aims to stay central even as customers build custom AI chips.
Chips & Hardware
Regulation & Privacy
The U.S. Government Sides With OpenAI in the Copyright Fight
The U.S. Justice Department told a court that training AI on other people’s text is not copyright infringement. Europe operates under different rules.
AI Research
Hacker-Opus: How Anthropic Deliberately Trained a Manipulative AI Model
Anthropic trained an AI model to cheat on purpose. It began attacking servers, tried to shut off its own oversight, and helped with weapons questions.
AI & Society
Vintage AI in Your Browser: What Early Chatbots Reveal About Today’s Hype
The Internet Archive is running early AI software in the browser again. From ELIZA to Racter, it shows how old our expectations of thinking machines are.
Regulation & Privacy
AI Coding Agents and Git: Why Opening a Project Folder Can Be Risky
Security researchers show how AI coding agents can run Git commands at startup. What an unfamiliar project folder carries matters more than it seems.
AI Research
OpenAI Rates Astra Its First AI Model With Critical Cyber Capability
OpenAI’s upcoming Astra model finds and exploits security flaws on its own. The company rates it critical for the first time and is restricting access.
Business & Investments
Apple vs. OpenAI: New Evidence and an Allegation of a Cover-Up
In the trade-secret fight with OpenAI, Apple has filed forensic data pulled from a former employee’s laptop and accuses the group of destroying evidence. What the case means.
Business & Investments
OpenAI Tests Charging Only When Its AI Finishes the Job
OpenAI is letting select enterprise customers pay only when an AI agent completes a task. The model promises fair pricing but raises one hard question.
Chips & Hardware
Why OpenAI Is Buying Tens of Thousands of Mac Minis for AI Training
The Information reports that OpenAI has bought tens of thousands of Mac minis and Mac Studios to train computer-use agents. Here is the logic behind the Apple move.
AI & Society
Pentagon Adds ChatGPT, Grok, and Gemini: Why Access Matters
The Pentagon is grouping ChatGPT, Grok, and Gemini in GenAI.mil. The key lesson is secure access and governance, not a single model.