Today In AI
The day's AI news from primary sources, refreshed every morning. Every story links to where it came from.
Anthropic Sets IPO Range of $70 to $78 per Share, Implying $1.404 Trillion Value
Anthropic priced its IPO at $70 to $78 per share. At the top of the range, the company would be worth $1.404 trillion on a fully diluted basis.
Why it matters: A listing at this scale shows how much capital markets now attach to frontier AI labs.
Source: The Verge · TechCrunch
Google Releases Gemini 4 Carbon to the Public with an Agent 'Swarm' Feature
Google is rolling out Gemini 4 Carbon to the general public. It first went to a select group of cybersecurity defenders two weeks ago.
Why it matters: A new top-tier model from Google changes which systems builders can pick from for hard tasks.
Source: Google DeepMind · The Verge
DeepMind's AlphaProof Code Verifies Signal's Core Encryption with 3 Million Lines of Proofs
AlphaProof Code agents wrote three million lines of proofs in Lean to verify Signal's core encryption. DeepMind says it works as intended.
Why it matters: Formal verification by AI could change how teams prove that critical software is safe.
Source: Google DeepMind · Wired
Put your product in the AI news feed
One sponsor card after the third story, labeled and matched to the day's topics.
See ad optionsOpenAI Launches Businesses in ChatGPT, Including Verified Business Agents
OpenAI is rolling out business profiles in ChatGPT. Verified Business Agents will let people chat with brands to do things like compare products.
Why it matters: ChatGPT is becoming a place where customers find businesses, so brands need to show up there.
Source: OpenAI · TechCrunch · The Verge
Anthropic Launches Claude Meetings as Part of Its New AI Office Suite
Claude Meetings turns workplace discussions into plans and tasks. Anthropic is expanding beyond its enterprise and coding focus.
Why it matters: Anthropic is moving into everyday office work, which puts it in more direct competition for team workflows.
Source: TechCrunch · The Verge
New York's AI Safety Law Takes Effect, Putting OpenAI and Anthropic Under Oversight
New York is preparing to enforce its AI safety law. Major developers must register, disclose safety practices and report critical incidents or face fines.
Why it matters: State-level rules are becoming real obligations for AI developers.
Source: TechCrunch
Nvidia Opens Probe into CEO Jensen Huang over Ties to Suspected Chinese Spy
Jensen Huang admitted an affair with Megaspeed exec Alice Huang. Nvidia's board opened an investigation, and the stock slid more than 4%.
Why it matters: Nvidia's chips and investments shape the AI industry, so leadership risk reaches every builder.
Source: Wired · The Verge · TechCrunch · Ars Technica
Anthropic Launches Claude Fable 5.2 and Mythos 5.2 with Faster Outputs
Anthropic released two new models it says improve coding, reasoning and vision, and generate outputs faster than the 5.1 versions.
Why it matters: Faster, stronger models for coding and reasoning can change what you ship with the Claude API.
Source: TechCrunch · The Verge
Anthropic Reportedly Presses Ahead with IPO and Buys $3.4B of Gas Turbines
The Wall Street Journal reports Anthropic is moving forward with IPO plans and may set a valuation range this week. It also bought $3.4B of natural gas turbines.
Why it matters: Anthropic's finances and power supply affect how reliably you can build on its models.
Source: TechCrunch
Meta Releases Muse Ember, a Cheap Lightweight Model for AI Agents
Meta's Muse Ember is built for agent execution and costs 10 cents per million input tokens, well below rivals like Haiku and Luna.
Why it matters: Agents make many model calls, so a low per-token price can sharply cut running costs.
Source: TechCrunch
Claude Code Prompt Is Accidentally Exposing Secrets on GitHub
Ars Technica reports that a "Do not ask again" prompt in Claude Code leads to passwords, API keys and other secrets being shared on GitHub.
Why it matters: A single permission click could leak credentials from your repos, so check your settings.
Source: Ars Technica
NVIDIA and Korea Expand Sovereign AI Work, from Chips to Startups and Models
NVIDIA and Korea's Ministry of Science and ICT are widening their partnership. It covers semiconductors, startups, open models and agent security.
Why it matters: It shows how nations are building local AI stacks, and which open Korean models are now easier to deploy.
Source: NVIDIA
Nvidia Opens Investigation into CEO Jensen Huang's Ties to Reported Chinese Spy
Nvidia's board is starting an independent investigation into CEO Jensen Huang's relationship with Megaspeed executive Alice Huang. Huang admitted to an affair in a memo to staff.
Why it matters: Leadership trust at the top AI chip maker matters to anyone whose stack depends on Nvidia.
Source: The Verge · TechCrunch
Apple Pushes to Speed up Siri's Shift into an AI Agent
Bloomberg's Mark Gurman reports Apple CEO John Ternus is pressing Craig Federighi to speed up the Siri roadmap. The goal is an AI agent that can compete with OpenAI's Dots or Meta's Muse.
Why it matters: Apple moving Siri toward agents could shape what users expect from assistants on their devices.
Source: The Verge
Elvis Presley and Eminem Rights Holders Sue AI Music Firm Suno
Rights holders of Elvis Presley's music and the publisher of some of Eminem's biggest records are suing Suno. They claim it illegally generates music and lyrics that mimic those artists.
Why it matters: Legal pressure on AI music tools could change how generative audio products are built and used.
Source: The Verge
NYT Cooking Adds Beta AI Tool to Swap Ingredients and Adjust Recipes
The New York Times has added a beta feature to its Cooking app that helps readers swap ingredients or adjust recipes. A Wired reviewer says it mostly hits the mark.
Why it matters: A publisher shipping a focused AI feature shows how narrow tools can add value to existing products.
Source: Wired
World Models Move AI from Text to 3D Space, Wired Explains
World models take AI beyond text into 3D space. They simulate real-world physics to teach robots and autonomous cars how to navigate reality, Wired reports.
Why it matters: World models could shape the next wave of robotics and autonomous vehicle tools.
Source: Wired
After the Midterms, Elon Musk and Data Centers Face a Rockier Road
TechCrunch says things are looking rocky for Elon Musk and data centers after the midterms. Its Equity podcast unpacked what the election means for the tech industry.
Why it matters: Data center politics can affect the cost and availability of the compute AI builders rely on.
Source: TechCrunch
METR Report: Anthropic Is Using AI to Build Better AI, and It Is Working
A new METR report says Anthropic's models are improving roughly twice as quickly as a year ago. The company is using AI to help build more powerful AI, and there is little sign of a slowdown.
Why it matters: If AI speeds up its own development, the pace of model upgrades you plan around may keep rising.
Source: The Verge
Nvidia H200 Chips Cleared for a Chinese Firm with an Empty Office
The New York Times reports that New Matrix is on the approved list to handle Nvidia's H200 AI chips in China, yet its offices are empty. Concerns are growing about how firms are vetted.
Why it matters: Weak vetting of chip buyers could invite tighter export rules that affect the AI hardware supply chain.
Source: The Verge
US Intelligence Memo Claims Megaspeed Executive Was Recruited to Spy on Nvidia CEO
A memo circulated earlier this year reportedly claimed China's Ministry of State Security recruited a female technology executive at Megaspeed to spy on Nvidia's CEO.
Why it matters: It shows how closely AI chip makers and their partners are being watched in the US-China tech contest.
Source: TechCrunch
Alex Stamos on How AI Has Left Cybersecurity in a Mess
Cognition's chief security officer Alex Stamos joins The Verge's podcast to discuss the state of cybersecurity. It is part one of a special series on AI in real life.
Why it matters: A veteran security voice weighing in on AI's impact is useful context for anyone shipping AI products.
Source: The Verge
NYT Cooking's AI Assistant Rescues a Weeknight Dinner but Nearly Ruins a Cake
A Verge writer tested the AI assistant in NYT Cooking. It helped with a weeknight chicken dinner, but she says it almost ruined a cake.
Why it matters: A real-world test shows AI helpers can be useful in one task and risky in another.
Source: The Verge
Caregiver Apps Add AI, but Caregiver Well-Being Is Still Unsolved
Apps that help family caregivers manage a loved one's needs are becoming more capable with AI. Wired says the wider issue of caregiver well-being remains unsolved.
Why it matters: It is a reminder that AI features help with tasks but may not fix the deeper problem a product serves.
Source: Wired
OpenAI Reportedly Raises $11 Billion More at a $1.25 Trillion Valuation
OpenAI has reportedly raised another $11 billion in a round valuing it at $1.25 trillion, according to Bloomberg. The round has not closed yet.
Why it matters: Capital at this scale shapes who can afford frontier compute, and how fast rivals must move.
Source: The Verge · TechCrunch
Voters Back Tighter Data Center Limits as Local Resistance Grows
In 40 of 41 data center ballot measures this week, a majority voted for more restrictions. Protesters also broke into a Microsoft site in Amsterdam.
Why it matters: Power, permits and public opinion are becoming real constraints on where AI capacity can be built.
Source: TechCrunch · The Verge
Free Chinese AI Models Close in on OpenAI's Cyber Model in Hacking Test
In an experiment with simulated networks, DeepSeek and GLM performed not far behind OpenAI's most advanced cyber model, which is not public.
Why it matters: Strong hacking skills in free models change the security baseline for anyone shipping software.
Source: Wired
Meta Adds Spaces to Muse, Letting the AI Build App-Like Experiences
Meta's new Spaces feature lets its Muse assistant build custom, app-like experiences you can use inside the Muse app.
Why it matters: Assistants that generate small apps on demand point to where consumer AI products are heading.
Source: TechCrunch · The Verge
Ramp Data Shows AI Spending Stalling Even as Usage Climbs
Ramp data shows AI usage still rising while spending has stalled. Price cuts and tighter cost controls at top spenders could test hypergrowth assumptions.
Why it matters: If big buyers are cutting AI costs, pricing and budgets may matter more than raw demand.
Source: TechCrunch
China Reportedly Recruited an AI Executive to Spy on Nvidia's Jensen Huang
The New York Times reports China's intelligence agency enlisted Megaspeed executive Alice Huang to gather information on Nvidia and its AI chips.
Why it matters: It shows how much the fight over advanced AI chips has become a matter of state intelligence.
Source: The Verge
OpenAI Drops NDAs with Officials for Data Centers After Midterm Backlash
OpenAI says it won't use NDAs with public officials on data center projects. It also unveiled a community compact, as Meta warns that voter pushback could make AI expansion harder.
Why it matters: Local opposition could slow the compute supply that AI products and startups depend on.
Source: OpenAI · The Verge · TechCrunch
OpenAI Launches GPT-6 Cyber, Built to Stay Within Authorized Scope
OpenAI's new cybersecurity model finds exploitable vulnerabilities and develops fixes. OpenAI says it went out of bounds less than 1 percent of the time in testing.
Why it matters: Security agents are only useful if they stay inside the limits you set for them.
Source: OpenAI · The Verge · TechCrunch
Claude Can Now Analyze Your Spending with a New Personal Finance Tool
Anthropic's Personal Finance feature connects Claude to bank and card accounts through Plaid. Users can ask questions like where their money went this month.
Why it matters: Chatbots are moving from answering questions to working with your private financial data.
Source: TechCrunch · The Verge
Perplexity Launches Always-On Teammates, Taking on OpenAI and Anthropic
Perplexity's new Teammates are always-on work agents that can take multiple roles in a workspace. They pull projects, skills and memory from tools users already have.
Why it matters: The race for workplace agents is widening beyond OpenAI and Anthropic.
Source: The Verge
Airbnb Plans an AI Travel Agent for 2027 as Uber Sees Few Agent Bookings
Airbnb says its AI agent will arrive in the first half of 2027 to plan and manage trips. Uber says only four people booked a ride through an AI agent in its busiest week last quarter.
Why it matters: Agent products face a gap between big plans and real consumer use today.
Source: The Verge · TechCrunch
Meta Adds Voice and Video Chat with Its Muse AI Avatar
Meta's Muse agent now supports real-time voice conversations. US users can also video chat with it as a customizable animated avatar.
Why it matters: Voice and video are becoming the default way to talk to AI agents.
Source: TechCrunch · The Verge