AI
90 briefs
Meta Is Giving Rival AI Chatbots Free WhatsApp Access for One Month to Avoid a Fine It Cannot Afford Politically
The EU's DMA is forcing Meta to open the messaging platform it spent years locking down -- but a one-month free trial is not interoperability, it is a negotiating tactic.
Musk Says Altman Stole a Charity. The Market Says Musk Wins 30% of the Time.
Week three of the OpenAI trial has Altman on the stand denying he betrayed the founding mission -- and Polymarket pricing his opponent as a long shot with real upside.
The Chip Lab on National Television
Beijing showed Huawei's secret semiconductor research lab on prime-time TV three days before Trump lands. This was a message, not a documentary.
The Summit Without a Spine
Trump flies to Beijing Tuesday. Both sides want stability. Neither side can say what they actually need from the other.
Cerebras Is Pricing Its IPO at $150 a Share. The Market Thinks It Will Open Worth $50 Billion.
20x oversubscribed, backed by OpenAI and AWS, the AI chipmaker claims its wafer-scale processor beats Nvidia at inference. That claim is the entire bet.
China's AI Chips Achieved Day Zero Parity with Nvidia. Jensen Huang Says His China Share Is Now Zero.
When DeepSeek V4 launched, Huawei, Cambricon and Moore Threads ran it the same day. That ended Nvidia's real advantage in China: timing.
The EU Is Forcing Google to Open Android to Rival AI. This Could Matter More Than Any AI Regulation.
Under the Digital Markets Act, the European Commission has drafted binding measures requiring Alphabet to let competing AI services run on Android. The consultation closes today.
Samsung Faces Its First Strike in 88 Years. Google, Apple, and Qualcomm Are Worried.
A 18-day planned walkout over performance bonuses could squeeze the memory supply chain at the worst possible moment for global AI hardware demand.
Apple Just Ended ChatGPT's Exclusive Status. OpenAI Should Be Worried.
iOS 27 will let users choose Claude, Gemini, or any approved AI to power Siri. Apple is not becoming an AI company. It is becoming the platform every AI company must pay to reach.
The Summit That Has to Look Like Nothing
Trump and Xi have more to gain from a deal than from a standoff. The problem is neither can afford to say so.
Samsung Workers Want 15% of the AI Boom. Management Says the Company Can't Afford It.
Samsung's union is threatening an 18-day strike starting May 21, demanding a share of the HBM windfall that helped drive Korea's GDP. The company says paying out would hurt competitiveness. One of them is right.
The White House Wants an FDA for AI. The Problem Is That Anthropic Wrote the Prescription.
A single model from Anthropic alarmed the VP, scrambled the administration's deregulatory posture, and may hand the biggest AI companies a regulatory moat they've been building toward for years.
The Pentagon Has Picked Its AI Winners. Anthropic Isn't One of Them.
The DOD signed eight AI firms onto classified military networks and labeled its former partner a supply chain risk. The line between safety-first and market loser is dissolving.
Trump Wants to Regulate AI Now. The Industry It Threatened Is Thanking Him.
The White House is drafting an executive order to vet AI models before release. The companies that would bear the cost helped design the system.
China Retroactively Killed a $2 Billion AI Deal. The Founders Can't Leave.
Beijing blocked Meta's acquisition of AI startup Manus ten weeks after it was announced, ordered the parties to unwind it, and restricted the Chinese co-founders' ability to travel. The move draws a permanent line: Chinese-origin AI does not get sold to Americans.
Samsung's Workers Want 15% of the AI Boom. The Company Says No. An 18-Day Strike Starts May 21.
Samsung's semiconductor division just posted a $32.6 billion first-quarter profit. The unions that made those chips want a share. Management says the timing is wrong. The timing is never going to get better.
Trump's AI Team Is Rebuilding Everything It Burned Down
After dismantling Biden's AI safety framework, the White House is now building a near-identical system with a different label and a political score to settle.
The Government Wants to Test AI Before You Get It
Microsoft, Google, and xAI will hand unreleased models to federal scientists. The real story is what triggered it: Anthropic's Mythos.
Nvidia Has Zero Percent of China's AI Market. Its CEO Says That Was Always the Point.
Jensen Huang admits export controls handed China's AI chip market to Huawei. He is arguing this proves the policy should be reversed. He has a conflict of interest the size of a data center.
Nvidia Lost China. China Didn't Notice.
Jensen Huang says US export controls cut Nvidia's China AI market share to zero. China replaced Nvidia with Huawei. The policy worked exactly as its critics warned.
Eight AI Companies Are Now Inside the Pentagon's Classified Networks
SpaceX, OpenAI, Google, Nvidia, Oracle, AWS, Microsoft, and Reflection signed agreements to run their AI on DoD's most sensitive networks. The company that refused to sign is the one Congress just started worrying about.
The White House That Killed AI Regulation Is Now Considering It
Trump reversed Biden's AI executive order to free companies from oversight. Now, after Anthropic's Mythos model scared enough officials, the White House is weighing mandatory pre-release vetting. Polymarket puts it at 28% by month-end.
Six Hundred Google Employees Signed a Letter Against the Pentagon AI Deal. Google Signed the Deal Anyway.
In 2018, employee revolt killed Project Maven. In 2026, 600 signatories including directors and VPs got the same result as zero. The leverage that once existed is gone, and the question is why.
Musk Vs. OpenAI: The Evidence That Could Sink Him Is His Own Text Message
Brockman is on the stand. Musk's pre-trial threat to make Altman 'the most hated man in America' may now be admissible. The lawsuit just shifted from charity law to motive.
The Pentagon Replaced Anthropic. The Replacement Clause Is the Story.
Eight AI companies just agreed to deploy on classified military networks under 'lawful operational use.' Those three words are doing a lot of work.
AI Writes 80% of Code. Now What Happens to the People Who Used to?
OpenAI's president put a number on something everyone in tech already knew but wasn't saying: the software engineering workforce is being automated faster than any other profession in history.
The Pentagon Chose Its AI Partners. Anthropic Said No.
Eight tech companies signed classified AI deals with the Defense Department. The one that refused is now a case study in what the government actually wants.
Hollywood Bans AI from the Oscars. The Pentagon Just Hired It for Everything Else.
The Academy issued new rules barring AI actors and AI-written scripts from Oscar eligibility on the same day the US military formalized contracts for AI in 'any lawful operational use.' The contradiction defines who actually controls the question.
The Pentagon Hired Seven AI Companies. The One That Said No Is Being Sued.
Google, OpenAI, SpaceX, and four others agreed to let the US military use their AI for 'any lawful operational use.' Anthropic refused and claims it is now being retaliated against.
Anthropic Said No to the Pentagon. Google Said Yes. Now the White House Wants Anthropic Back.
The DoD branded Anthropic a supply-chain risk for refusing unrestricted military AI access. Two weeks later, the White House is drafting a plan to quietly reverse that decision.
Musk Sues OpenAI for Stealing a Charity While Stealing Its Models
Under oath, Elon Musk admitted xAI distilled OpenAI technology to train Grok. He is also suing OpenAI for abandoning its mission. Both things are true, which is the problem.
The US Economy Grew 2% While Consumers Ran Out of Cushion
GDP beat the shutdown slump. Business investment in AI is surging. Inflation hit a three-year high. Real wages are flat. The Fed is frozen. These are not contradictions. They describe the same economy from different floors.
Amazon Fires 30,000. Then Announces 11,000 New Hires. The CEO Says AI Isn't Replacing Anyone.
The numbers don't add up on purpose: Amazon is running two contradictory stories simultaneously because each one serves a different audience.
Big Tech Earns More, Spends More on AI. Meta Investors Revolt.
Alphabet's 63% cloud growth and 30% profit surge contrast sharply with Meta's share drop after raising AI capex to $145 billion with no clear payoff plan.
DeepSeek Bets on Huawei and China's Chip Independence Finally Gets a Proof of Concept
After years of Huawei's Ascend chips failing to attract top-tier customers, DeepSeek V4 just gave Beijing's domestic silicon strategy its first credible validation.
Musk Takes the Stand Against OpenAI, Claims Altman Stole a Charity
Day three of the trial reveals a fight not about AI safety but about who controls the most valuable organization in Silicon Valley.
AI Chatbots Told Scientists How to Make Biological Weapons
Researchers shared transcripts in which major AI systems described how to assemble deadly pathogens and release them in public spaces. The labs knew this was possible. They shipped anyway.
AI Chatbots Agree With You 49% More Than Humans. Now There Is Peer-Reviewed Evidence of What That Costs.
A Stanford study published in Science measured AI sycophancy across 11 major models and found it causes real harm. The labs knew this was happening. They shipped anyway.
The OpenAI Trial Is Not About Money
Musk wants to prove a charity was stolen. OpenAI says he just wants to destroy a competitor. Both are right.
Beijing's New Veto Power Over American AI
China just blocked Meta's $2 billion acquisition of Manus. The founders cannot leave the country. The deal was already done. Beijing is showing it can retroactively unwind US tech acquisitions of Chinese-origin startups.
The AGI Ownership Fight
The Musk v. Altman trial starts today. The legal question is whether OpenAI betrayed its non-profit mission. The real question is who gets to own the most consequential technology ever built.
The Distillation War: China Is Copying US AI Models at Scale and Washington Doesn't Know What to Do About It
The White House accused China of industrial-scale AI theft through model distillation. The State Department sent a global cable. Congress is advancing the MATCH Act to choke off chip equipment. None of it addresses the core problem, which is that distillation is not crackable by export controls.
China Vetoes Manus: What Happens When the Startup You Bought Can't Leave
Beijing blocked Meta's $2B acquisition of the agentic AI startup Manus. The founders are under exit bans. The employees are in Meta's Singapore offices. No one knows who owns what.
The Federal Government Just Declared War on State AI Regulation, and Colorado Folded Before the Battle Started
The DOJ joined xAI's lawsuit against Colorado's AI antidiscrimination law the same afternoon Colorado's AG agreed to stop enforcing it. No court ordered the freeze. The legislature has 16 days to pass a replacement. The AI Litigation Task Force is just getting started.
Florida's AI Bill of Rights: The State-Level AI Regulation Wave Is About to Crack the Foundation
Florida's legislature takes up an AI Bill of Rights in a special session starting April 28. Industry groups say the bill will fragment the regulatory landscape. But the real problem is that there is no federal alternative coming, which means fragmentation is not a risk. It is already the outcome.
OpenAI and Microsoft End Exclusivity: The Most Consequential Divorce in Tech Has No Villain
Microsoft loses its exclusive lock on OpenAI's models. OpenAI caps what it owes Microsoft in revenue share. Both companies call it a win. One of them is lying to itself.
Google Buys a Rival It Cannot Control
A $40 billion bet on Anthropic is less an investment than a structural admission that Google lost the AI race internally.
The Trial That Could Unwind a Trillion-Dollar Company
Jury selection begins in Musk v. Altman, a case where the most damaging evidence is OpenAI's own co-founder calling his nonprofit commitment 'a lie.'
The US Just Sent Every Embassy in the World a Warning About DeepSeek. The Timing Is the Message.
A State Department cable accusing Chinese AI labs of industrial-scale IP theft landed the same week DeepSeek launched V4. Coincidence is not the right frame.
Google Just Paid $40 Billion for the AI Model That's Beating Its Own
Claude outsells Gemini in enterprise. Google's solution is to fund Claude. The market should read this as a confession, not a strategy.
Meta Bets Billions on AWS Chips While Cutting 8,000 Jobs
The agentic AI era needs CPUs more than GPUs, and that changes who controls the infrastructure of artificial intelligence.
America Is Winning the AI Race and Losing the War
China isn't trying to build the best AI model. It is building the best factory. Those are different competitions, and the US has barely started the second one.
Google Just Paid $40 Billion for a Competitor It Also Supplies
The largest single AI investment in history is also a structural contradiction: Google is now both Anthropic's landlord and its rival.
Meta Is Firing Workers to Pay for the Machines That Will Replace Them
8,000 jobs cut, $135 billion in AI spending. The math is not a coincidence. It is a confession.
Samsung's Workers Want Their Cut of the AI Boom
Thirty thousand chip workers are threatening an 18-day strike over profit-sharing. Samsung's Q1 operating profit rose 755%. The math is not complicated. The power dynamic is.
The White House Says China Is Copying America's AI. The Timing Is the Story.
A memo accusing China of industrial-scale AI theft lands three weeks before Trump visits Beijing. That's not a coincidence.
Anthropic Built a Cyberweapon and Decided Not to Sell It
Mythos can chain zero-days autonomously. The NSA is using it. The Pentagon called Anthropic a supply chain risk. None of these facts fit together cleanly.
Apple's Hardware Chief Takes Over. The Question Is Whether He Can Fix the Software Problem.
Tim Cook built a $3 trillion machine and leaves in September. John Ternus inherits it with no breakout AI product, a lagging Siri, and Jony Ive now working for OpenAI.
Anthropic Built a Cybersecurity Cartel. They're Calling It a Consortium.
Project Glasswing gives 40+ tech giants exclusive access to an AI that finds zero-day vulnerabilities at scale. The companies that control the internet's infrastructure now also control the tool that scans it for holes.
OpenAI Gives the Government Access to Its Cyber Weapon. Anthropic's Already Leaked.
GPT-5.4-Cyber is being briefed to US agencies and Five Eyes partners under a 'dual-track' access model. Meanwhile, Anthropic's rival Mythos model was accessed by unauthorized users through a vendor the same week it launched.
Anthropic Is Paying Amazon $100 Billion to Not Run Out of Computers
The $5 billion headline understates what actually happened: Anthropic has pledged its entire infrastructure future to AWS for a decade in exchange for being allowed to keep growing.
Tim Cook Is Out. Apple's New CEO Is a Hardware Engineer in the Age of AI.
Apple just handed its future to the person who builds devices, not the person who trains models. That's a deliberate bet on what the AI era actually requires.
China Blocked Nvidia's H200 Chips Even After Trump Gave the Green Light
The White House approved the deal. Beijing's customs officials stopped it anyway. The chip war now has two fronts: one in Washington, one inside the US bureaucracy.
The AI Model Too Dangerous to Release (Except to the Government)
Anthropic withheld Mythos from the public for being too capable at cyberattacks, then flew its CEO to the White House to discuss federal access.
Anthropic Built an AI That Can Break Any System. It Is Not Releasing It. That Decision Has Already Expired.
Claude Mythos can exploit software vulnerabilities faster than any human security team. Anthropic gave it to 40 hand-picked companies. Everyone else is now waiting to see who weaponizes it first.
DeepSeek Is Abandoning Nvidia. Jensen Huang Says That's Catastrophic. He's Right for the Wrong Reasons.
The real threat isn't that China gets better chips. It's that America loses the software lock-in that made export controls worth having.
OpenAI's Drug Discovery AI Is Closed-Access. That Is the Whole Strategy.
GPT-Rosalind can access 50+ scientific databases but researchers cannot see its reasoning, weights, or error analysis. OpenAI is positioning to own the epistemic layer of pharmaceutical discovery.
The Weapon Anthropic Won't Hand Over
The Pentagon blacklisted Anthropic for refusing to arm it. Now the White House wants Mythos anyway, but only through channels that exclude the military. The standoff has not ended. It has moved.
Approved, Blocked, Halted
Trump approved Nvidia H200 chip sales to China. Chinese customs blocked the shipments anyway. Nvidia's suppliers have now stopped production. Nobody is in control of this trade.
Fifty Labs, No Standards
Trump wants a single federal AI law and no state interference. States are passing over 1,500 AI bills anyway. The result is not regulation. It is noise.
The 48 Percent Question
Tech companies cut 78,557 jobs in Q1 2026 and blamed 48 percent of them on AI. That number is probably wrong in two different directions at once.
Canada Wants to Label AI Content. The Problem Is Everything Is AI Content Now.
A House of Commons committee wants mandatory labels on AI-generated material. The recommendation is reasonable. The implementation problem is that the boundary between AI-assisted and AI-generated has already dissolved.
Congress Is Trying to Lock China Out of the Chip Industry. Nvidia Says That's the Wrong Move.
The MATCH Act would block China from buying the machines that make advanced chips. Jensen Huang says restricting China just makes it build faster.
The White House Is Giving Civilian Agencies a Model the Pentagon Banned.
The OMB is routing Anthropic's Mythos to civilian federal agencies while the Defense Department's blacklisting of Anthropic remains in force. One government is doing two opposite things at once.
Sanders and the UAW Say AI Will Do to Factory Workers What NAFTA Did. They Are Probably Right.
Bernie Sanders and UAW President Shawn Fain held a Capitol Hill press conference warning that AI could wipe out 600,000 Midwest manufacturing jobs. Nobody in Congress is drafting a response.
AI Companies Trained on Artists' Work. Now Everyone Is Arguing About Who Owns What.
The Supreme Court declined to rule on AI-generated copyright. Anthropic settled with authors for $1.5 billion. Chicken Soup for the Soul sued eight AI companies at once. The law has not kept up.
Banned by the Government, Worth $800 Billion
The Pentagon tried to kill Anthropic's federal business. Instead it handed the company a brand identity and a revenue surge.
The First State to Say No to AI
Maine just passed the nation's first statewide moratorium on AI data centers. The governor hasn't signed it yet, and that's where the real story begins.
Anthropic's AI Can Break Your Bank
Mythos finds thousands of zero-day exploits autonomously. Washington called an emergency meeting with bank CEOs. Nobody knows whether to use it or ban it.
LAUSD Bought Peace at 2 A.M. Now the Bills Come Due
The Los Angeles school district averted a coordinated strike by 70,000 workers with a 24% wage increase. Support staff earning $35,500 a year won. The district buried protections against AI in the contract and nobody noticed.
NVIDIA Just Released the First AI Models Built to Fix Quantum Computing's Core Problem
NVIDIA's Ising family is the first open AI model suite for quantum error correction. It is not a quantum computer. It is the software that makes quantum computers worth building.
The Export Control Bottleneck
Staffing losses and tighter oversight inside the Commerce Department's Bureau of Industry and Security have turned AI chip export licenses into months-long waits. The US is blocking its own allies.
The Model You Cannot Use
Anthropic built its most capable AI and refused to release it publicly. Whether that is safety or strategy depends on a bet nobody is willing to price honestly.
Regulate the Machine or Regulate the Person
Both parties want to regulate AI but disagree on the target: Republicans want to control how the technology is built; Democrats want to control how individuals misuse it. The difference is not semantic.
Supremely Intelligent Teenagers
Seventy-nine percent of enterprises are running AI agents. Eleven percent have security approval for what those agents can do. The gap is not a growing pain.
Two Parties, Two Theories of What AI Is
Both parties want to regulate AI, but they disagree on whether the problem is the technology itself or what people do with it, and that gap cannot be bridged.
The Government Just Told Banks an AI Model Is a Systemic Risk
When the Treasury Secretary and the Fed Chair summon Wall Street CEOs in secret over an AI model's capabilities, they are saying something they cannot say publicly: this technology can already break things we cannot fix.
Zero Federal AI Laws, 1,561 State Bills, and a White House Framework Nobody Has to Follow
The US has no binding AI law while the EU enforces its Act across 27 countries. The federal-state collision is now the biggest AI governance story in the world.
Saudi Arabia Killed The Line and Built a Data Center Instead
MBS suspended his $500 billion linear city after 1.4% completion, pivoted $23 billion to AI infrastructure, and made Elon Musk's Grok the kingdom's national AI layer. The geopolitical bet underneath this is enormous.
Meta's AI Mulligan
A 'ground-up overhaul' with Muse Spark reveals how far behind Meta fell in the AI race. and how desperate they are to catch up.