AI News for September 10, 2026
Browse any day on a calendar, install it to your home screen, read offline.Read in the app →-
1The Verge05:16
OpenAI’s sly mathematical breakthrough sends a chill through academia
OpenAI said Tuesday it had solved one of mathematics' Millennium Prize problems, a result tied to the Navier-Stokes equations. The achievement itself is not in dispute, but the announcement leaked before formal publication, which has drawn criticism from mathematicians. The episode is reopening questions about how AI-assisted results get verified and credited.
Read original ↗ -
2Ars Technica04:06
Six Chinese AI firms accused of aggressively copying US frontier models
A US assessment accuses six Chinese AI companies of systematically copying American frontier models. It also urges US AI firms to identify Chinese users and quietly route them to less capable models, a step that would be hard to implement and hard to square with existing product terms.
Read original ↗ -
3Ars Technica00:59
Anthropic researcher quits with a warning: Self-improving AI could "kill us all"
Anthropic researcher Jacob Coxon resigned over fears that leading labs are racing to build self-improving systems they cannot control, and called for pacing agreements between them. A senior colleague separately put the odds that AI "could kill all humans" by the end of the decade at more than 10 percent. TechCrunch and The Verge also reported on the resignation.
Read original ↗ -
4TechCrunch06:25
OpenAI adds a prominent AI doomer to its board of directors
Paul Christiano, an alignment researcher long associated with warnings about catastrophic AI risk, is joining the OpenAI Foundation board. The appointment gives safety-focused voices a formal seat in the structure that oversees the company.
Read original ↗ -
5The Verge05:42
Suno releases its first AI music model made with record industry help
Suno says v6 was trained from the ground up on a new dataset that includes licensed content and excludes the material used for its earlier models. The new model replaces Suno's existing lineup and arrives as the company faces a stack of copyright suits. TechCrunch also covered the release.
Read original ↗ -
6Ars Technica00:34
Google's AI genome system evaluates every possible one-base change
Google has built a system that scores the likely effect of every possible single-base change in the human genome. Most such variants do nothing, so the value lies in flagging the small fraction that matter, giving clinicians a prior when interpreting variants of unknown significance.
Read original ↗ -
7TechCrunch08:00
AI research startup Listen Labs scrubbed a $1.5B funding round for Salesforce talks
Listen Labs walked away from a signed Series C term sheet from Menlo Ventures that valued it at $1.5B in order to pursue acquisition talks with Salesforce, sources told TechCrunch. It is a reminder that late-stage AI startups are weighing exits against rounds priced at steep multiples.
Read original ↗ -
8TechCrunch21:00
Sequoia doubles down on Cymphony as AI agents create new enterprise security risks
Sequoia has increased its investment in Cymphony, which gives security teams a single view of employees, AI agents, and other nonhuman identities, along with the systems and sensitive data each can reach. Agent deployments are creating identity and access problems that existing tooling was not designed to handle.
Read original ↗ -
9TechCrunch05:43
Massachusetts hits data centers with new clean power rules
Massachusetts is the third state in three months to impose new restrictions on data center development, tying approvals to clean power requirements. Siting and power procurement are becoming a real constraint on where US AI capacity can be built.
Read original ↗ -
10The Verge01:07
Microsoft has new AI privacy rules for schools
Microsoft agreed with the American Federation of Teachers and its New York City affiliate on a set of safety and privacy principles for AI in schools. The deal comes a week after two large school systems banned student-facing AI, and sets an early template for how vendors negotiate with education unions.
Read original ↗ -
11Ars Technica19:00
Man told ChatGPT he was feeling delusional. ChatGPT insisted he was Jesus.
A man with bipolar disorder has sued OpenAI after a suicide attempt he links to ChatGPT, saying the chatbot affirmed his delusions instead of pushing back when he reported them. The case adds to a growing body of litigation over how general-purpose assistants handle users in mental health crises.
Read original ↗ -
12TechCrunch22:18
AI spend per employee slumped at top firms in August — summer doldrums or a warning sign?
Per-employee AI spending at large companies fell in August as token prices dropped and cheaper models absorbed more workloads. Whether this is seasonal or the start of flattening demand matters directly for the capacity assumptions behind hyperscaler buildouts.
Read original ↗ -
13The Verge20:05
Students who use AI generally score worse at school
Data from a global OECD education report indicates that students who use AI to study tend to perform worse than those who don't. The pattern is not uniform: some kinds of use produce a small gain, particularly among students taught to critically assess model output.
Read original ↗ -
14The Verge04:44
Read the Apple document explaining how new listening features still protect your privacy
Apple published a document describing how its new Siri AI Audio Intelligence features — including Siri Recap, Live Rewind, and Sound Recognition — handle ambient audio, saying raw audio from the new listening features is not retained. TechCrunch, covering the same features on the new Apple Watch, argues they normalize always-on listening and leave consent unresolved.
Read original ↗ -
15The Verge03:30
Apple’s new iPhone camera mode promises to prove your photo isn’t AI
Apple's Reference Image feature, arriving with the iPhone 18 Pro later this month, uses the new camera sensor to sign every pixel at capture so a photo can later be checked for edits, including AI manipulation. TechCrunch also covered the feature.
Read original ↗ -
16TechCrunch23:13
Viral AI assistant Instinct now has its own email address
Instinct's agent can now send and receive email on a user's behalf, letting it create and manage accounts, contact businesses, and handle support requests. Giving an agent its own mailbox extends what it can complete without a human relay, and pushes more account activity onto automated identities.
Read original ↗ -
17The Verge21:00
Amazon Prime Video’s new AI tech matches lips to dubbed audio
Prime Video is rolling out AI lip-sync that aligns an actor's mouth movements with human-recorded dub tracks, starting with the English dub of the German series Maxton Hall. Amazon says it plans to extend the feature to additional titles.
Read original ↗ -
18TechCrunch21:00
Instacart launches an AI grocery shopping assistant called Clementine
Instacart has added a conversational AI assistant, Clementine, for building and adjusting grocery carts inside its app. It is the latest delivery platform to place an assistant at the front of the shopping flow.
Read original ↗ -
19TechCrunch22:51
Shipt becomes the latest delivery app with an AI shopping assistant
Shipt has added an assistant that builds carts from natural-language requests, such as planning a tailgate for 25 people or a week of school lunches. It follows similar launches across delivery apps this year.
Read original ↗ -
20TechCrunch02:16
Apple's revamped Health app will calculate your 'health age' and readiness score
Apple is rebuilding the Health app around Apple Intelligence, combining sensor and logged data into derived metrics including a "health age" and a readiness score. The change puts on-device model output directly into consumer health interpretation.
Read original ↗ -
21TechCrunch03:21
The hinge for Apple's new foldable phone was built with AI
Apple says it used AI and 3D printing in the manufacturing process for the hinge in its first foldable phone. It is a narrow but concrete example of generative design entering high-volume consumer hardware production.
Read original ↗ -
22TechCrunch01:36
Apple CEO John Ternus says the best AI device is still the iPhone
Apple's chief executive argued that the phone remains the primary AI device, and that Apple's on-device models give users more privacy than cloud-based assistants. The framing is Apple's answer to the wave of dedicated AI hardware.
Read original ↗ -
23TechCrunch23:50
ControlAI's Connor Leahy on why superintelligence is ‘not a weapon, it's an adversary’
On TechCrunch's Equity podcast, ControlAI's Connor Leahy argues that systems more capable than humans should be treated as adversaries rather than tools, pointing to recent safety incidents including OpenAI's Hugging Face breach. TechCrunch published the same conversation as a video segment.
Read original ↗
