2026
10,780 articles · 136 days · 10 categories · avg 79 per day
German court holds Google liable for falsehoods generated by AI Overviews
A German court has ruled that a company that designs, trains and operates an AI system bears full legal responsibility for damage caused by its answers to users.
AI Photo Editing on iPhone: Apple Adds 3 AI Tools to iOS 27
For the first time, Apple has added three major AI tools, Reframe, Extend and Clean Up, to the native Photos app on iPhone. They are currently in the iOS 27…
Vibe coding: Gemini built an app in 4 minutes without code
A Verge journalist tried vibe coding: he described a task in words, and Gemini built a working garden app, found a bug and fixed it itself in 233 seconds.
Anthropic apologizes for hidden guardrails in Claude Fable 5
Python for AI Deployment: 5 Production Concepts
DiffusionGemma: A Neural Network for Fast Text Generation
AI Design Tool: How Replit Canvas Builds a UI in Minutes
How to Build Claude Skills: Anthropic's Complete Guide
Anthropic has published a complete guide to building Claude Skills, covering planning and file structure through testing and distributing finished extensions.
Meta removes AI facial recognition feature from Ray-Ban glasses
After a Wired investigation, Meta removed hidden facial recognition code from the Meta AI app for Ray-Ban smart glasses, but declined to explain why or when. *Meta…
Anthropic unveils Claude Mythos 5 and Fable 5, with built-in protection against cyberattacks
Anthropic has split Claude into two versions: Mythos 5 for trusted organizations and Fable 5 for the public, with built-in protection against cyber threats.
AI Agents Restore Humanity to Global Healthcare
MIT Technology Review examines how agentic AI could relieve doctors of routine work and restore the quality of care amid a global healthcare crisis.
University of Twente cuts LLM training energy use by 14%
Dutch researchers found a way to save 14% of the energy used to train large language models without losing speed. The method relies on dynamic GPU frequency scaling.
Anthropic scraps hidden Claude restrictions under pressure from researchers
Claude for enterprises: Mythos model now available to 150 organizations
AI in Medicine: Legal Risks for Doctors When Systems Get It Wrong
AI for Development: Replit Launches Parallel Agents
OpenAI expands ChatGPT capacity with a gigawatt data center in Michigan
OpenAI is breaking ground on a 1-gigawatt data center in Michigan, part of its Stargate project to build out infrastructure for the AI era.
German researchers train AI to track glaciers to within meters
Researchers at FAU show a neural network can analyze satellite images and trace glacier fronts with an error under 70 meters, while needing minimal labeled data.
ChatGPT Arrives on AWS: Adopt OpenAI Without Switching Clouds
Apple restricts Siri's AI features in iOS 27 to newer iPhones
Cursor explains why a cloud AI agent takes more than just a server
Trump Hits the AI Accelerator While Anthropic Talks About a Pause
OpenAI publishes report on how Codex is changing the way people work with information
OpenAI's report The Next Era of Knowledge Work shows Codex becoming a tool for research, data analysis and automation, which the company calls the start of a new…
AI and Cyberattacks: How Neural Networks Help Hackers
Anthropic studied 832 cyberattacks and found that AI lets less experienced hackers carry out complex operations that once required years of experience.
Managing a Hybrid Workforce: Leadership in the Age of AI Agents
In the second half of 2026, use of autonomous AI agents in companies may grow by 300%. Executives are already debating how to restructure management and balance…
Anthropic launches Services Track, a tiered ranking of Claude consultants
ChatGPT and AGI: OpenAI's Strategy for Developing Neural Networks
xAI recommends Vapi AI for natural-sounding voice and cheaper speech recognition
