AI News
AI news that matters. Updated daily.
No stories match your filters.
Fields Medal Winner Timothy Gowers Says AI Solved a Major Open Math Problem
"AI has now solved a major open problem - one that many mathematicians had tried." That's Timothy Gowers this week, and the source matters as much as the claim.
Spotify and Universal Music Let Fans Make AI Covers, Artists Get a Cut
For years, AI-generated music was a legal minefield. Record labels sued over unauthorized voice clones, streaming platforms quietly removed AI tracks, and artists watched their likenesses get replicated without consent or payment. Spotify and Universal Music Group just drew a new line.
AI Name-Reading System Skips Hundreds of Graduates at Commencement Ceremony
Hundreds of graduates walked across a stage this spring to silence - or worse, the wrong name - after a college deployed an AI system to handle commencement announcements and it allegedly missed hundreds of students entirely. The crowd responded with boos.
Google Search Is Going Full AI - 6 Alternatives That Still Show You Links
Three years ago, a Google search meant ten blue links. Today, an AI-generated summary sits above everything else on most informational queries, and Google is pushing further - toward a mode where the results page itself becomes a conversation rather than a directory. TechCrunch flagged the shift this week with a list of six alternatives worth bookmarking.
Anthropic Drops 13 Free Claude Courses - Certifications Incoming
Anthropic released 13 free official courses with certificates, covering Claude development from basic API usage up through agentic workflows and MCP (Model Context Protocol - the open standard Anthropic built for connecting AI models to external tools and data sources like databases, file systems, and APIs).
Trump Delays AI Security Review Order, Citing Risk to U.S. AI Lead
President Trump declined to sign an executive order that would have required AI companies to submit their models for government security review before public release. He told reporters he was dissatisfied with the order's language and didn't want regulation to interfere with the U.S. maintaining its lead in AI development.
xAI Becomes First Major AI Lab to Open Its Books via SpaceX IPO Filing
The major generative AI companies have operated as financial black boxes since they launched. SpaceX's IPO filing this week changes that for xAI, making Elon Musk's AI company the first of the big generative AI labs to disclose its finances through a public document.
Anthropic Launches 13+ Free AI Courses with Official Certificates
Anthropic just made its entire AI training curriculum free, releasing 13+ courses that each end with an official certificate from the company. The catalog covers everything from prompt engineering basics to Claudee Code(/tools/claude-code/) (Anthropic's terminal-based coding assistant) to building agentic AI systems - AI that takes sequences of actions autonomously, like browsing the web or writing and running code, rather than just answering a single question.
Spotify and UMG Strike Deal for AI-Generated Remixes as Premium Add-On
Spotify and Universal Music Group announced a licensing deal that will let Premium subscribers generate AI-powered remixes and covers of catalog songs. The feature ships as a paid add-on on top of a standard Premium subscription, with exact pricing not yet disclosed.
Google's Gemini Can Clone Your Face on Video. The Results Are Unsettling.
The first time you see an AI video of yourself that you didn't actually make, the reaction isn't wonder. It's something closer to discomfort.
Spotify Studio App Generates AI Podcasts from Your Email and Calendar
Three months into the wave of AI-generated daily briefings, Spotify is entering from an angle none of them can match: your listening history.
Spotify Adds ElevenLabs AI Narration to Its Author Self-Publishing Platform
700,000 audiobook titles. That's how many Spotify already hosts - and now the platform wants to help authors create more of them without hiring a narrator.
Spotify Launches AI Q&A and On-Demand Podcast Briefings for Premium Users
What if your podcast app could tell you whether an episode is worth 90 minutes of your time before you press play?
Spotify's New Desktop App Takes Direct Aim at NotebookLM for Personal Podcasts
Last year, turning a folder of research into a listenable podcast meant exporting from Google's NotebookLM, loading it into a separate app, and syncing it to your phone. Spotify just cut all of that out.
Google Puts Ads Inside AI Mode, Changing How Businesses Get Found in Search
Google is putting ads inside AI Mode, its conversational search experience that lets users ask multi-turn questions and get synthesized answers from across the web. The format, announced by Google, lets businesses engage with users directly through real-time questions and interactions - closer to a chat interface than a list of sponsored links.
Meta Sends Legal Notice to Heretic, an Open-Source LLM Project
Meta has served a legal notice to the Heretic Free Software Project, a small open-source initiative working with local large language models - AI models run on your own hardware rather than accessed through a cloud service.
At Anthropic's London Developer Event, AI-Written Pull Requests Were the Opening Icebreaker
Three months into Claudee Code(/tools/claude-code/)'s commercial availability, Anthropic held a two-day developer conference in London. Code with Claude kicked off on May 19, the same day Google staged its I/O developer event in Palo Alto. Anthropic staffers said the scheduling overlap was coincidental.
Brett Adcock's Hark Raises $700M at $6B Valuation for Secretive AI Interface
$700 million. That's what investors poured into Hark, Brett Adcock's new AI startup, in a Series A round that values the company at $6 billion - before it has publicly shown what it actually builds.
New AI Therapy App Claims 95 on Mental Health Safety Benchmark vs. 65 for Consumer Chatbots
The best consumer AI chatbots score 65 on Vera-MH, a benchmark that tests whether AI handles mental health conversations safely. The Path, a new AI therapy startup, claims its model scores 95.
Musk's OpenAI Lawsuit Ends in Loss After a Month of Courtroom Theater
After roughly a month of daily protests outside the courthouse, celebrity-level press scrums, and the spectacle of two of tech's most prominent figures in the same room, Elon Musk's lawsuit against Sam Altman and OpenAI ended without the result Musk was seeking.
Google's AI Agent Push Has a Consumer Problem
What happens when a platform company decides AI agents are the future, but the people it needs to convince haven't asked for them?
Anthropic's $15B/Year SpaceX Deal Revealed in IPO Filing
$15 billion a year. That's what Anthropic is paying SpaceX for access to its Colossus data centers in Memphis, Tennessee, according to details in SpaceX's S-1 IPO filing.
148 Words Into a Browser, 10 Minutes Later: A Working Android App
148 words. That's all it took to ship a working Android app to a physical phone - no IDE setup, no Gradle configuration, no Android SDK installed. Just a browser tab and Google's AI Studio.
AdventHealth Deploys ChatGPT for Healthcare Across 50-Hospital System
Nurses at AdventHealth now have an AI assistant handling their documentation. The 50-hospital system has deployed ChatGPT for Healthcare - OpenAI's HIPAA-compliant version of its assistant built specifically for medical environments - to reduce administrative workload and give clinical staff more time with patients.
Google Confirms Ads Are Coming to AI Mode Search Results
Google's AI Mode - the search experience that gives you a detailed AI-generated answer instead of a list of blue links - is getting ads. Google confirmed the change at Google Marketing Live on May 21, according to a post on the Google Ads blog.
The Slop Grenade: Why Pasting Raw AI Output Into Conversations Backfires
Someone drops a 1,400-word AI response into the Slack thread. It has six headers, three bullet lists, and opens with "Certainly! Here's a comprehensive overview..." Nobody reads it. The question that prompted it needed two sentences.
Meta Cuts Thousands of Jobs to Fund Its AI Infrastructure Push
Meta is cutting thousands of employees, according to reporting from The Verge, framing the reductions as a cost-efficiency move needed to sustain its AI investment commitments. In an email to affected staff, management described the layoffs as part of a "continued effort to run the company more efficiently."
OpenAI Claims Its Model Solved an 80-Year-Old Math Problem
OpenAI is claiming a mathematical breakthrough: one of its models reportedly solved an open problem that professional mathematicians have been working on for 80 years. The company has not yet released a peer-reviewed paper, which means the proof hasn't been independently verified. Until it is, this is a claim worth taking seriously but not one to treat as settled fact.
Developers Run Head-to-Head Tests: Claude Code vs Copilot vs Opencode on Qwen3 27B
Developers are increasingly running the same coding task across multiple AI tools to see where the real differences are - and a recent comparison pitting GitHubb Copilot(/tools/github-copilot/), Claudee Code(/tools/claude-code/), and opencode against each other using Qwen3 27B is a good example of what that looks like in practice.
Claude Code User Chronicles Building an MVP in 7 Days from a Standing Start
Seven days. That's the timeline one developer documented this week going from opening Claudee Code(/tools/claude-code/) for the first time to a working product they're calling, with obvious hyperbole, a unicorn.
Google Rebrands Vertex AI as the Gemini Enterprise Agent Platform
Google is restructuring its enterprise AI offering under a new brand called the Gemini Enterprise Agent Platform, effectively replacing Vertex AI as the primary face of its cloud AI product line.
Claude Code's Product Lead on Usage Limits, the "Lean Harness" Design
Developers who hit Claudee Code(/tools/claude-code/)'s usage limits without warning have had a consistent complaint: the limits are real but the feedback isn't. Anthropic's product lead for Claude Code addressed this directly this week, acknowledging the transparency gap and outlining how the team thinks about capacity constraints on what has become one of the most actively used AI coding environments available.
Judge Delays Approval of Anthropic's $1.5B Copyright Settlement
A federal judge has put the brakes on Anthropic's $1.5 billion copyright settlement, declining to give it immediate approval and adding a new layer of legal uncertainty to one of the largest AI training-data cases on record.
SpaceX IPO Filing Names Grok's 'Spicy' Mode as a Litigation Risk
$500 million. That's how much SpaceX has set aside for potential litigation losses in its IPO filing - and one of the named risks is Grok's "Spicy" mode.
Intuit Cuts 3,000+ Jobs to Shift Spending Toward AI
3,000 jobs. That's the number of positions Intuit announced it's eliminating as the company redirects its workforce and budget toward AI across its product portfolio.
Jensen Huang Bets $200B on CPUs for AI Agents as Nvidia's Next Hardware Play
$200 billion. That's the size of the CPU market Jensen Huang says he's found inside the AI agent boom - and he's betting Nvidia can own it.
Anthropic Tells Investors Q2 Will Be Its First Profitable Quarter, With $10.9B Revenue
For most of its short life, Anthropic has been a money-burning research lab kept alive by billions from Amazon and Google. That's about to change.
What Claude's Visible Thinking Actually Signals - and Why Users Have Learned to Dread It
What happens when you can watch an AI model think? Apparently, you start dreading certain phrases.
Meta Details How It Scales Safety Testing for More Capable AI
Meta published a blog post this week on how its internal development and testing processes are changing as its AI systems grow more powerful and more personalized. The central argument is straightforward: a model that adapts to individual users is harder to test exhaustively than one that behaves the same way for everyone, and that gap compounds as capability increases.