Artificial intelligence is more a part of our lives than ever before. While some might call it hype and compare it to NFTs or 3D TVs, generative AI is causing a sea change in nearly every part of the technology industry. OpenAI’s ChatGPT is still the best-known AI chatbot around, but with Google pushing Gemini, Microsoft building Copilot, and Apple adding its Intelligence to Siri, AI is probably going to be in the spotlight for a very long time. At The Verge, we’re exploring what might be possible with AI — and a lot of the bad stuff AI does, too.
Co-founder Jack Clark, who will lead the new Anthropic Institute, said he had “no concerns” about research funding.
Now users in Canada, New Zealand, and India can access Chrome’s built-in Gemini AI assistant, which has added support for more than 50 languages, including Spanish, French, Hindi, and Chinese.
Along with answering questions about what’s on your screen, Gemini in Chrome can help you do things like send messages in Gmail, create a table comparing products in your tabs, and remix images you see online.
Bluesky user @lifewinning.com just coined a catchy term for Grammarly’s AI agent masquerading as Verge staff members. After backlash last week, Superhuman says writers can now email them to opt out of Grammarly’s “expert review” feature.
“A little offended Grammarly didn’t make a sloppelganger of me.”
On Tuesday, Amazon expanded access to its Health AI platform beyond just One Medical members to include Amazon.com and its app, alongside an introductory offer for Prime members.
Similar to ChatGPT for Healthcare, Amazon says its Health AI is a HIPAA-compliant tool to answer general health questions, analyze medical records, and connect users to medical professionals through One Medical.
Coco Robotics, which makes autonomous delivery robots, announced on Tuesday that it’s partnering with Niantic Spatial, the AI spinoff of the Pokémon Go developer. Niantic’s geospatial AI model, which was trained on images from AR scans in Pokémon Go, will help Coco’s robots find their way to your house.
Amazon’s eCommerce SVP, Dave Treadwell, called an all-hands meeting on Tuesday to address recent outages linked to AI coding agent errors, the Financial Times reports. That includes more oversight around AI coding, with Treadwell announcing that “junior and mid-level engineers will now require more senior engineers to sign off on any AI-assisted changes.”
[Financial Times]
They joined a new initiative called Utilize that aims to use strategies like battery storage and virtual power plants to make more use of the electrons already available to the grid. It’s a plan that’s supposed to make electricity more affordable as opposition grows to data centers blamed for higher utility bills.
The AI startup, founded by ex-OpenAI executive Mira Murati, will team up with Nvidia on a “long-term gigawatt scale strategic partnership” to power TML’s AI model training. The news comes after multiple founding members of the startup left for OpenAI at the same time earlier this year.
[Thinking Machines Lab]
During an interview with Bloomberg, Emil Michael, the under secretary of defense for research and engineering, says Anthropic’s move to sue the Pentagon was an “expected reaction” to its designation as a supply-chain risk.
“I don’t think there’s a scenario where this gets resolved in that way,” Michael says, Bloomberg reports.
His Paris-based startup, Advance Machine Intelligence, just raised $1 billion to build AI world models.
It’s another big bet on what might be the next big thing in AI: systems trained on the physical world, not just text and images.
A new toggle is being rolled out in Google Photos that allows users to choose between “classic” and AI-powered “Ask Photos” modes when searching through the app. The AI search had a rocky launch, and now Google Photos lead Shimrit Ben-Yair says users want more control over their results.
An empty tome listing nearly 10,000 authors — including Kazuo Ishiguro — calls out “theft” by AI companies as the UK government weighs sweeping changes to copyright law. It follows last year’s “silent album” stunt.
Organizers are handing out copies at today’s London Book Fair. Just don’t let Anthropic shred them.
Experienced white-collar workers are now part of a miserable gig economy.
The results from a new March 2026 NBC News poll of 1,000 registered voters showed the only two items on their list with lower negative net ratings than AI’s 26 percent positive / 46 percent negative split were Democrats and Iran.
About 56 percent said they had used an AI platform like ChatGPT or Copilot in the previous month.
The tool, which OpenAI said is used by 25 percent of Fortune 500 companies, helps “identify and remediate vulnerabilities in AI systems during development.”
[OpenAI]
The multi-agent tool, called Code Review, should catch “bugs human reviewers often miss,” Anthropic said. Agents run in parallel and deliver a high-level overview, plus in-line comments for individual issues.
Code Review is available in research preview for Enterprise and Teams customers.
After acquiring Arduino last October Qualcomm has announced a new single-board computer called the Ventuno Q. Pricing and availability aren’t known, but it will be powered by a Dragonwing IQ8 processor, 16GB of RAM, and a 40 TOPS NPU. It’s designed for building robots and machines that function autonomously in response to input from connected sensors.
The Cowork integration was built in close collaboration with Anthropic and aims to help Copilot perform “long-running, multi-step tasks,” according to Microsoft’s announcement. The feature is in testing and will be available to preview later this month through Microsoft’s Frontier program.
ChatGPT’s long-promised “adult mode” is still on the way, but will miss its planned Q1 launch. For some, that means the long, lonely wait continues.
pretendworld:
reading the news that your new girlfriend has been delayed must be just brutal
Get the day’s best comment and more in my free newsletter, The Verge Daily.
Elon Musk’s short-lived agency rolled into the NEH with the mandate to cancel grants that it deemed contrary to Donald Trump’s anti-DEI agenda. According to the New York Times, decisions about which grants to cancel weren’t made after careful analysis and deliberation. Instead, they were made with a ChatGPT prompt.
… instead of looking closely at funded projects, they pulled short summaries off the internet and fed them into the A.I. chatbot.
The prompt was simple: “Does the following relate at all to D.E.I.? Respond factually in less than 120 characters. Begin with ‘Yes’ or ‘No.’” The results were sweeping, and sometimes bizarre.
[New York Times]
Caitlin Kalinowski posted on X that she resigned from OpenAI, saying the company’s contract didn’t do enough to protect Americans from warrantless surveillance and that granting AI “lethal autonomy without human authorization” was a line that “deserved more deliberation.”
The open source tool poses plenty of risks, but for devotees, it’s an antidote to Big AI.
The feature was expected to launch sometime this quarter, but an OpenAI spokesperson tells former Verge staffer Alex Heath that “We’re pushing out the launch of adult mode so we can focus on work that is a higher priority for more users right now, including gains in intelligence, personality improvements, personalization, and making the experience more proactive.”
OpenAI just launched Codex Security, a new research preview AI agent focused on identifying and fixing app security issues. Separately, it announced the Codex Open Source Fund is also now including “conditional access” to Codex Security as part of the six-month ChatGPT Pro with Codex subscriptions it’s offering open source developers.
All I can say is in the brief time I used the Cluely app, it helped me cheat on exactly nothing. So I am not surprised to see this TechCrunch report alleging that CEO Roy Lee lied about the company’s revenue numbers.
