News

The AI and technology stories that matter, as they break: the deals, launches and decisions reshaping the field.

LatestLLMs

DeepMind alumni's AI agent beats Claude and GPT-5.5 at redoing science

Inherent, founded by Google DeepMind alumni, unveiled its AI research agent Faraday after emerging from stealth in August 2026

EthicsAugust 24, 2026

Why training AI on copyrighted books lands in legal gray zone

Fair use, a 1976-era copyright doctrine, is being applied to AI training on hundreds of millions of books and articles with mixed results in court.

Is it legal to train AI models on copyrighted books? It’s complicated | TechCrunch

LLMsAugust 24, 2026

Mystery AI model Ox Alpha sparks guessing game over its maker

A stealth model called Ox Alpha appeared on OpenRouter on August 22, 2026, listed by an anonymous third-party provider

PolicyAugust 24, 2026

Egypt sets EGP 138.9bn ICT investment target, banks on AI push

Egypt targets EGP 138.9 billion in ICT investment for FY 2026/27, with 83.1% expected from private and foreign capital.

Gov’t targets EGP 138.9 bn in ICT investment, mostly from private capital

AI AgentsAugust 23, 2026

Torvalds credits AI in Linux commit after brutal Xe driver bug hunt

Linus Torvalds credited an AI tool in a Linux kernel commit message fixing a drm/xe (Intel graphics) driver bug

AI AgentsAugust 23, 2026

Harvard's $699 Startup Bootcamp Adds AI Avatars of Its Mentors

Harvard's HBS Foundry, an $699, eight-week startup bootcamp, now offers AI avatars of its instructors, built by startup HeyGen.

A collage of the five 2026-2027 Blavatnik Fellows

AI AgentsAugust 23, 2026

Why Reviewing Every Line of AI-Generated Code Falls Short

Simon Willison's August 2026 post argues that manually reading every line of AI-generated code doesn't scale and isn't the most effective way to catch problems.

Top 7 Code Review Best Practices For Developers in 2025

EthicsAugust 23, 2026

OpenAI Tops, Meta and Anthropic Trail in AI Containment Study

Guidelight AI Standards graded Anthropic, Google, OpenAI, Meta, and xAI on their plans for containing a rogue AI model.

These AI Models Can’t Stop Breaking Out Of Their Cages

AI AgentsAugust 23, 2026

LinkedIn's 'AI slop' button has been clicked over a million times

LinkedIn's 'Seems like AI slop' flagging button has been clicked over one million times since its July 30th launch

HardwareAugust 23, 2026

Starcloud raises $250M as it bets on orbital AI data centers

Starcloud raised a $250M Series A extension, valuing the orbital AI data center startup at $2.3 billion

AI AgentsAugust 23, 2026

Google Research Deploys AI Agents to Vet Wearable Health Biomarkers

Google Research unveiled the Biomarker Discovery Framework, a multi-agent system that mines wearable sensor data for candidate health biomarkers.

CoDaS: AI Co-Data-Scientist for Biomarker Discovery via Wearable Sensors - Daily Arxiv - 안광섭 ex)haebom

LLMsAugust 23, 2026

Simple Ranking Beats RL for Picking AI Explanations, Study Finds

A new arXiv paper finds LambdaRank, a pairwise ranking method, beats PPO, GRPO, DPO and other selectors at choosing offline AI-generated explanations

AI AgentsAugust 23, 2026

Meta's Pocket lets US users vibe-code games from a text prompt

Meta launched Pocket to US users on August 20, 2026, letting people generate playable games called 'gizmos' from text prompts

Meta just launched a vibe-coding app for games, and it’s called Pocket

EthicsAugust 23, 2026

Pew Study: 35% of Web Pages Since ChatGPT Show AI Authorship

Pew Research found 35% of English-language web pages published since ChatGPT's November 2022 launch show signs of AI authorship, using Open Pangram's detection technology.

A Third of New Websites Are AI-Generated

LLMsAugust 20, 2026

Google Bundles Study Tools Into a New Gemini Hub for Students

Google launched a student hub in Gemini combining study notebooks, flashcards, and quizzes, alongside new Search-based study tools (August 2026)

AI EthicsAugust 20, 2026

OpenAI Launches ChatGPT for Teens With Study Mode, Quiet Hours

OpenAI launched ChatGPT for Teens on August 18, 2026, for users aged 13-17, with automatic enrollment based on account age or estimated age.

Editorial image for OpenAI Launches ChatGPT for Teens With Parental Controls

LLMsAugust 19, 2026

Black Forest Labs' Flux 3 Adds Native Audio to AI Video

Black Forest Labs launched Flux 3, its first model to generate video with native, synchronized audio, in clips up to 20 seconds long.

HardwareAugust 19, 2026

Amazon's Texas AI data center could become the US's top polluter

Amazon's planned Pecos County, Texas data center includes an on-site natural gas plant permitted to emit 33 million tons of CO2 annually.

Planned Amazon data center could become the biggest climate polluter in the U.S.

All articles