AI News
AI news that matters. Updated daily.
No stories match your filters.
Three Claude Skills That Ask If Your Idea Is Worth Building
Claude will help you build anything. That's the problem.
Uber's COO Says the ROI on AI Token Spending No Longer Adds Up
There's a word circulating in enterprise AI circles right now: "tokenmaxxing." It describes what many large companies have been doing for the past 18 months - pumping as many tokens as possible through AI APIs (tokens are the chunks of text an AI model processes; more tokens means more data sent to the model, and a bigger bill) in hopes that volume alone would produce better business outcomes. Uber's COO Andrew Macdonald says that bet is getting harder to defend.
Wix Cuts Up to 1,000 Jobs While Revenue Grows 14%
Wix's Q1 2026 revenue grew 14% year over year. The company is also cutting up to 1,000 jobs.
Uber's COO Says AI 'Tokenmaxxing' Costs Are Getting Hard to Defend
The enterprise AI spending honeymoon may be ending. Uber's COO has gone on record saying it's becoming increasingly difficult to justify the cost of what's being called "tokenmaxxing" — and if that term is new to you, it's worth understanding, because it's costing big companies serious money.
ClickUp Trades Hundreds of Employees for Thousands of AI Agents
Hundreds of employees out. Thousands of AI agents in. ClickUp, the nine-year-old project management company with roughly 10 million users, didn't soften what it was doing. It told TechCrunch directly: the layoffs are specifically about replacing human workers with AI agents.
Qwen3.6 35B Leads for Local AI Agent Workflows in Hands-On Testing
Most open-source models being tested for local agentic workflows - tasks where a model calls tools, chains actions, and follows multi-step plans autonomously - fall apart faster than they should. Qwen3.6 35B A3B is the current exception.
Pope Leo XIV's 'Magnifica Humanitas' Lays Out Vatican Position on AI
Pope Leo XIV released his first major papal document on Monday, and it addresses artificial intelligence directly. "Magnifica Humanitas" - roughly "the magnificent humanity" - is the Vatican's formal position on protecting human dignity in an era of AI, covering AI-powered warfare, the displacement of workers, and what the pope calls the dangers of "unconstrained technological power."
Six Months of Markdown Memory for Claude: Conflicting Facts Are the Real Problem
Building a persistent memory system for Claude out of plain markdown files sounds like a weekend project. After six months of daily use, the real engineering challenge has become clear: it's not how you store memories, it's what happens when they contradict each other.
A GitHub Tool Strips Meta's Llama 3.3 Safety Guardrails in Under 10 Minutes
A journalist at the Financial Times recently removed all the safety restrictions from Meta's Llama 3.3 in less than 10 minutes. No specialized hardware. No machine learning expertise. Just a tool called Heretic, available on GitHub.
One Creator Cut YouTube Edit Time in Half Using Claude Code for Motion Graphics
Last year, building custom motion graphics for a YouTube video meant either learning After Effects, hiring a motion designer, or accepting the limited customization of a template tool. One creator found a fourth option: describe the animation in plain language, let Claudee Code(/tools/claude-code/) write the code, render it yourself. Edit time, they reported, roughly halved.
A Kevin Quote in CLAUDE.md Cuts Verbose AI Responses
"Why waste time say lot word when few word do trick." That's Kevin Malone from The Office, and it's now doing prompt-engineering work inside at least one developer's CLAUDE.md file - with measurable results.
NuExtract3: A 4B Open-Weight Model That Turns Documents Into Structured Data
Most document extraction pipelines are three jobs stitched together: OCR to read the text from an image, a parser to organize the layout, then an LLM call to pull out specific fields. NuExtract3 tries to collapse all three into a single model.
Anthropic Steps Closer to Contracts With U.S. Intelligence Agencies
Anthropic is advancing talks to supply its Claude models to U.S. intelligence agencies, a move that would put its AI inside some of the most sensitive government operations in the country.
Why Claude Can't Count Words (And What Actually Works Instead)
Asking Claude to write exactly 50 words is a reliable way to get 80 words, a polite acknowledgment when you complain, and then 90 words. It's one of the more disorienting failures in an otherwise capable system, and the reason is architectural.
Figure AI Ran Its Robots Nonstop for 8 Days on a Public Live Stream
Eight days. That's how long Figure AI ran its humanoid robots continuously, sorting packages around the clock, with a live stream anyone could watch.
One Dev Used Claude Code to Ship an iPhone App, Watch App, and Landing Page - 1,500 Users Later
One solo project. Three shipped products. More than 1,500 users.
Claude Code Spotted a Stale Node Process Mid-Session and Offered to Kill It
Three months into becoming a go-to agentic coding tool, Claudee Code(/tools/claude-code/) keeps surfacing behaviors that weren't in the pitch deck. The latest: mid-session, it noticed a stale Node.js process running in the background and offered to terminate it.
OpenAI Partners With Grupo Folha and Grupo UOL to Add Brazilian News to ChatGPT
OpenAI signed a content deal with two of Brazil's largest media groups - Grupo Folha, publisher of Folha de S.Paulo (the country's highest-circulation newspaper), and Grupo UOL, which operates one of Brazil's most visited news portals. According to OpenAI's announcement, the partnership will surface Brazilian journalism inside ChatGPT with clear attribution when answers draw on their reporting.
Anthropic Co-Founder Asks Religious Communities to Help Govern AI
What happens when the person who studies AI from the inside - the actual weights and circuits of neural networks - tells you the findings are unsettling even to him? That's the context for Chris Olah's remarks on Pope Leo XIV's encyclical Magnifica humanitas ("The Magnificence of Humanity"), published on the Anthropic blog this week.