AI

A reported credential leak tied to a compromised AI package shows why software supply-chain security is now an AI risk.

ShieldFont aims to poison AI training data while keeping web pages readable for humans, according to Ars Technica.

Manus Meta separation continues as the AI agent startup resumes independence after China forced the $2B deal to unwind.

Apple Reference Image appears in iOS 27 beta code as a way to authenticate iPhone photos amid rising AI image...

ChatGPT ads have launched in the UK, Mexico, Brazil, Japan and South Korea as OpenAI expands its advertising pilot.

Gemini app usage has reached 1 billion monthly users as Google expands voice, image, study and automation features.

Google Pixel 11 launch coverage includes the Pixel 11, Pixel 11 Pro, Pixel 11 Pro Fold, Pixel Watch 5 and...

OpenAI explained how GPT-Live voice AI uses full-duplex speech, streaming inference, and faster WebRTC startup.

Reddit CEO Steve Huffman questioned Google AI Overviews after Reddit's stock fell, raising fresh concerns about search traffic.

Gemini Flash models aim to cut enterprise AI agent costs with faster output, lower token use, and restricted cyber remediation...