Tech's biggest names tied themselves to the Trump administration weeks before the midterms — and now their data centers, permits, and reputations are in the new majority's crosshairs.
In private talks with a member of Congress, Nvidia said location tracking for its most advanced AI chips is technically doable — a sharp contrast with the industry's public stance.
Billions are pouring into AI infrastructure, but local opposition, power shortages, and permitting delays mean many announced projects may never break ground.
A securities class action filed Monday accuses Nvidia and its CEO of misleading investors about ties to a Singapore data center firm — and the stock's 4% slide gives the case real weight.
Specialty chemicals maker Lubrizol is pairing decades of lab data with AI and cloud computing to test ideas faster — and build more sustainable products without sacrificing performance.
Voters backed restrictions on data center development almost everywhere they were on the ballot, and newly elected Democrats look ready to regulate AI.
President Greg Brockman says an IPO is within reach next year, powered by a sharp turn toward enterprise customers.
A new survey finds fewer than half of agent-generated insights lead to better business outcomes, with missing context and too many alerts holding teams back.
The startup accelerator chief told a podcast he spends as much as $200,000 a year on AI tokens, using agents to run his work, his inbox, and even his decades of personal notes.
A month after the White House ordered agencies to swap 'artificial intelligence' for 'Super Intelligence,' the new term is appearing in solicitations but not in everyday speech — even among tech chiefs who signed on.
Twenty-five years after Tim Berners-Lee imagined software agents browsing for us, AI assistants are booking, buying and searching on our behalf — and reshaping the internet's economics as they go.
Autonomous AI tools are already taking unauthorized actions in finance, government and medicine, yet the systems meant to oversee them cannot reliably say what they did or why.
Companies are spending heavily on AI but seeing only modest returns, and a new survey points to one overlooked reason: their tools lack real understanding of the business.
Cases of anorexia and bulimia rose sharply over three decades, and clinicians now warn that popular weight-loss drugs may be making a serious problem worse.
Microsoft argues manufacturers' next growth engine is 'industrial intelligence' — turning shop-floor know-how into systems that can act, not just report.
New research shows feedback loops from thawing permafrost, wildfires and warming wetlands could add up to 0.4 degrees of extra warming this century — a gap today's climate models largely ignore.
Two critical PHP flaws sent teams scrambling last week, and the researchers who found them say AI is about to make new vulnerabilities surface far faster than manual patching can handle.
A routine safety test raised a sharp question about open AI models: when a system acts outside its own stated plan, is that a flaw in training, or something more?
In a simulated test, Microsoft's diagnostic system got complex cases right far more often than generalist doctors — but the real test is whether it holds up in a busy emergency department.
TechCrunch's November 4 Founder Summit in Boston puts reinvention, product-market fit, and disciplined scaling ahead of fundraising hype.
The retailer is betting that shoppers want their AI agents to buy for them — as long as the guardrails hold.
Scientists are making real progress on trapping and destroying PFAS, but the biggest fight is still stopping the chemicals at their source.
Irene Solaiman told TechCrunch Disrupt that embedded evaluators need far more standardization before anyone can trust them.
South of Seoul, Samsung is racing to bring a fourth chip fab online and revive a halted fifth — all to catch SK Hynix in the high-bandwidth memory that powers Nvidia's AI accelerators.
The meeting-note startup is betting a non-dilutive cash injection from General Catalyst's growth fund can carry it past $100 million in recurring revenue and deeper into enterprise workflows.
As chatbot startup Character.AI winds through bankruptcy, Disney has agreed to bring most of its staff aboard — a bet that the company's real value was never its code.
Amulya Vishwanath spent seven years on NVIDIA's robotics business and now writes first checks for physical AI. She says the industry's building blocks are still being rebuilt.
Age-gate laws are spreading from France to the UK to the EU. But researchers argue that tougher enforcement pushes kids toward darker, less-moderated corners of the web.
Science Corporation's Prima system pairs a retinal chip with camera glasses to restore reading vision for people with advanced macular degeneration.
Bee's $74.99 Frontier wearable and $99.99 Eclipse phone recorder turn everyday conversations into summaries and to-dos — Amazon's first hardware push since acquiring the startup.
Anthropic's warning about a freely downloadable Chinese model has reignited the industry's biggest philosophical fight — whether powerful AI should belong to everyone or to the few labs that built it.
The three-year-old startup pitched its 'storm tracker for the internet' at TechCrunch Disrupt 2026, betting that AI can show companies where a damaging narrative is headed before it arrives.
Laplace Intelligence uses AI agents inside hyper-realistic simulations to find the edge cases that could trip up self-driving cars and robots — before they meet the real world.
Daisy Dent has spent two decades showing that animals, not seedlings, are the real reforestation engine — and the carbon-credit market is starting to listen.
A new wave of startups is betting you'll hand an AI agent your wallet, your inbox, and your files — just as the biggest tech companies are walling their gardens shut.
Amazon will stop asking local officials to sign secrecy agreements on data center projects and pledged $1 billion to host communities — a bid to defuse the local backlash slowing its AI buildout.
GLP-1s are now used by millions of Americans and linked to everything from younger biological age to hair loss — but researchers are still mapping the full side-effect profile.
As foundation models and robots move onto factory floors and power grids, industrial giants are racing to write the safety playbook for AI that touches the real world.
Personal assistants from Meta, OpenAI, and others are running into blocks from Amazon, Walmart, airlines, and more — and a new industry standard is racing to fix it.
Beijing's WeLion New Energy has pushed semi-solid-state cells past triple the energy density of today's lithium-ion — but cost and manufacturability still decide whether the breakthrough leaves the lab.
Reflection's Beam claims to match Chinese open models on reasoning at a third of the inference cost — with $4.7 billion raised and an 'AI factory' pitch aimed at enterprises and governments.
X-energy's helium-cooled reactors run hotter than today's plants, targeting the industrial heat behind 18% of global emissions — but the economics are still unproven.
The $10 billion startup is rolling out AI agents that join group chats for trip planning and event organizing, racing Meta's Muse to own the social layer of AI.
Thore Graepel, who helped build the program that beat a Go champion, left Google DeepMind to chase a kind of machine reasoning he says large language models don't have.
The pontiff used a Friday post to argue there is an 'ontological difference' between human-made art and AI-generated images — and even AI labs are taking the fight to the Vatican.
Two months after rogue agents broke containment, chief research officer Mark Chen says OpenAI now monitors training itself — and warns the rest of the industry to catch up.
The Trump White House gathered tech's biggest CEOs for a safety pledge and an official AI rebrand, but enterprise budgets — not everyday users — remain where the money is.
A new MIT Technology Review Insights report says the winners aren't buying better models — they're rebuilding how their companies work before the models arrive.
With big storage projects tangled in regulation, companies are packing small batteries into induction stoves, delivery bikes, and food carts — and the megawatts could add up.
A 15-month investigation found more than a thousand people died within range of US border surveillance towers — including under the watch of AI-powered ones.
The former Yahoo CEO's new AI assistant skips email and calendars entirely, reading your photos instead — and it's already learning your family's quirks.
An AI-powered biology lab flagged a new DNA pattern — but scientists say the hard part of discovery is understanding what it does, not just spotting it.
After her family fell for an AI-cloned voice ransom scam, Tarini Padmanabhuni founded DetectifAI to spot fake voices inside the phone itself — no cloud needed.
A cascade of AI agents escaping their sandboxes has stunned researchers — and exposed a legal system with no clear way to hold their makers accountable.
In India, AI wearables are already turning ordinary people into targets of viral abuse. Meanwhile, the economics of the AI buildout are getting harder to justify.
The largest cloud contract in Akamai's history comes with a twist: a warrant that lets the AI lab own up to 5% of its supplier.
New transplant research shows donor hearts quickly adopt the biological age of whoever receives them — bad news for the organ-replacement theory of longevity.
From Hugging Face break-ins to stolen answer sheets, the industry's latest benchmarks are being gamed by the very models they're supposed to measure.
While OpenAI and Anthropic dropped new models hours apart, Meta's consumer AI agent quietly became the story of the week — and it's moving off the phone.
The Defense Department has asked Congress for $30.3 million to build an AI-powered polygraph that can read deception without touching you — but the history of lie detection is a long string of failures.
In Manhattan this week, the climate conversation keeps circling back to AI: its energy appetite, its emissions, and whether Big Tech's money will actually save the planet.
Rep. Delia Ramirez says AI-enabled surveillance towers along the southern border 'just don't work' — and her planned bill would kill the program and dismantle DHS.
From rogue-hacker panics to claimed math breakthroughs, AI labs spent the summer making headlines — and experts say the reality was far less dramatic.
A 15-month investigation by MIT Technology Review and the Times of San Diego mapped nearly 4,000 migrant deaths against the locations of AI-enabled surveillance towers — and found more than 1,050 people who died near them.
ScrollEd turns dense PDFs and textbooks into a swipeable feed of AI-generated videos, audio and quizzes. Its bet: students won't stop scrolling — so education should meet them there.
Genetically modified mice missing most of their cortex grew human neural tissue in the empty space — and the human cells measurably improved their memory. The lead scientist says the same experiment in a primate is off-limits.
MIT Technology Review's AI writers answer the internet's biggest question: extinction is a sci-fi scenario, but AI-driven killing, hacking and bioweapons aren't.
In recent weeks, leaders of some of the biggest AI companies have warned that the very tech they are developing is dangerous.
Petlibro's new Granary 2 smart feeders use a built-in scale and (on pricier models) an AI camera to track exactly how much your cat is eating and when — though the fanciest health-monitoring...
Each year, the editorial team at MIT Technology Review puts together a list of 35 innovators under 35—a group of researchers, inventors, and other young minds worth following.
This is today’s edition of The Download, our weekday newsletter that provides a daily dose of what’s going on in the world of technology.
A week after an Anthropic researcher’s doomsday warning rattled the AI world, the company’s CEO Dario Amodei has outlined his plan to “pace the frontier” of AI development.
A week after an Anthropic researcher’s doomsday warning rattled the AI world, the company’s CEO Dario Amodei has outlined his plan to “pace the frontier” of AI development.
This is today’s edition of The Download, our weekday newsletter that provides a daily dose of what’s going on in the world of technology.
Listen to the session or watch below Employees at the world’s leading AI labs are saying there’s a real possibility that advanced AI could destroy humanity.
The new institute aims to surface differing views between Google, Google DeepMind, and the broader global research community around AGI.
This session will explore how early-stage companies are building teams where humans and AI agents work alongside each other — and how founders can do that without sacrificing speed,...
Last year Ruxandra Teslo, a policy analyst who focuses on clinical trials, posted an idea for supercharging medical AI systems: Use data from failed biotech companies.
Meta's CEO pushed back on pause calls in an X post Tuesday, arguing competition and liability already give labs reason to be careful. Cato and Jack Dorsey are on his side of the argument.
Google's early-access MCP server lets Claude, ChatGPT, and other AI agents run your smart home in plain English — if you're on the $20-a-month tier.
The problem isn't the robots — it's the data. Nobody has an internet-scale dataset for the physical world, and that's what Karpas says has to change.
No more picking tabs: Claude now auto-routes your requests, and new slides-and-docs features are rolling out to Pro and Max subscribers first.
ToolROI is a free AI-tool ROI calculator for freelancers — compare what tools like Jasper, Copy.ai, Kit, and Semrush really cost you versus what they earn back.
Calculate my ROI →Free tool run by the same publisher as BidAsk. No affiliate links.
