Google renamed NotebookLM to Gemini Notebook. The branding catches the product up with the rest of Google’s AI portfolio, but you’ll be more interested in some of the new features.

The UK government has announced plans to make social media unavailable to 16- and 17-year-olds from midnight to 6 a.m. by default, and will also disable infinite scroll and autoplay. A separate planned package would interrupt extended AI chatbot sessions for users under 18.

New York Gov. Kathy Hochul has imposed a one-year moratorium on new hyperscale data centers that require more than 50 megawatts of electricity. It is the first statewide moratorium of its kind in the country, and a serious strategic mistake.

OpenAI has launched ChatGPT Work, a new desktop experience designed to do more than answer questions. It can work across your files, applications, browser tabs, and connected accounts to complete multistep assignments.

In my Sunday essay, I argued that the United States is winning a race China isn't actually running in. America is building the best models while China is building the models that are "good enough." Will "good enough" prove to be more valuable than best?

That is a remarkable product transition. Claude is beginning to evaluate how you think, what you delegate, and which parts of your intellectual life you retain for yourself.

Is this a fight over which superpower will hold the keys to the world’s AI economy? If so, the question is: who is the rabbit, and who is the fox?

Anthropic has launched Claude Cowork on mobile and web. Cowork sessions now run in the cloud by default, so tasks keep working after you close your laptop, and scheduled tasks run with no device online at all.

Particle 6 announced “Misaligned,” the first feature film for Tilly Norwood, the AI “actor” that enraged Hollywood in late 2025 after reports that talent agencies were interested in representing her.

Today, a small cluster of frontier labs (Google, OpenAI, Anthropic, and a few well-funded challengers) control access to the best AI models. Imagine a world where that is no longer true. Would that world be different in kind, or just different in degree? Let's explore.

Anthropic has released Claude Science, a research workbench that pulls the tools a scientist juggles all day into one place where an agent runs the analysis and a second agent checks its citations and math. The product is exciting, but let’s explore the dark side.

On June 30, the Commerce Department lifted the export controls that had forced Anthropic to switch off Claude Fable 5 and Mythos 5. Today, Fable 5 returns. The models went dark for eighteen days, and the reason they went dark deserves your attention.

TIDAL announced yesterday that starting on July 15, fully AI-generated music on the platform will no longer earn royalties. The company will tag those tracks with an “AI” badge, remove anything that impersonates an artist, and block 100% AI material from direct-to-fan sales.

Here's a question: If models can write most prompts now, what exactly is the human's job? Three senior engineers at Anthropic, Google, and OpenAI all arrived at the same answer independently: "loop engineering."

OpenAI will release GPT-5.6 as a limited preview to a small group of enterprise customers at the request of the Trump administration, with the federal government approving access on a customer-by-customer basis during the preview window.

The Associated Press reports that Anthropic’s Mythos model, running red-team exercises with U.S. intelligence agencies under an Anthropic initiative called Project Glasswing, identified vulnerabilities in highly sensitive classified U.S. government computer systems.

Anthropic has released Claude Tag in beta.

Queries about real people, real events, and real places benefit from images of those actual people, events, and places. Trust in AI search depends on the image being a real record of a real moment. Is this the next phase of editorial content monetization?

Greetings from Cannes Lions. AI will be at the front and center of almost every conversation this week, and many (if not most) of these talks will focus on exactly the wrong thing. My Sunday essay explores what we can learn from how businesses deployed a previous world-changing technology.

I'm getting ready to head to Cannes, and a lot of the AI conversations are going to focus on enterprise deployment at scale. I’m expecting the following four areas to be displayed and debated in every conceivable way. I wrote this piece for Think with Google. Have a look.

Anthropic CEO Dario Amodei spent Wednesday afternoon in a closed-door G7 working lunch in Évian-les-Bains asking President Trump to lead an international AI coalition.

AI deployment at enterprise-scale begins and ends with your corporate knowledge base (by far the highest hill to climb). Where and how to run it just got a little easier. You can now choose from any of the big three hyperscalers.

U.K. prime minister Keir Starmer announced Monday that the United Kingdom will ban social media for everyone under sixteen, naming Snapchat, TikTok, YouTube, Instagram, Facebook, and X.

On Friday, Anthropic received a U.S. government directive to suspend access to Fable 5 and Mythos 5 for any foreign national, inside or outside the U.S., including foreign-national Anthropic employees. My Sunday essay explores the fact that the era of permissioned intelligence has arrived.

Wall Street is already looking toward OpenAI, Anthropic, and the next wave of AI listings. Today’s IPO may set the comps for the entire AI class of 2026.

If this legal logic travels, operators of any AI that publishes a defamatory sentence may become defendants.

If you’re debating a “rent the brain” partnership, this week made the asymmetry visible.

Maybe, possibly, there’s a chance that someday soon, Siri won’t suck.

When you buy electricity, you buy kilowatt-hours. When you buy bandwidth, you buy bits-per-second. When you buy storage, you buy gigabytes. When you buy AI intelligence in 2026, you buy tokens, which are easy to meter, easy to price, but hard to value. This sounds like a technical detail. It isn’t.

How will you create a culture of continuous adaptation so that your organization does not fall behind?

If you depend on referral traffic, Google has now confirmed where search is headed. Plan accordingly.

Microsoft got into the Corporate Claw business yesterday by announcing Scout: a personal AI assistant built on the OpenClaw framework and connected to Microsoft 365.

Anthropic is the first major frontier AI company to publicly announce that it has begun the IPO process. More importantly, Anthropic is a Public Benefit Corporation.

My Sunday essay explores the value of building synthetic ideological adversaries and how to use them. AI is not a substitute for a human structured argument (devil's advocate, Socratic method, dialectic); it's an amplifier that helps you ask "The Next Question."

Anthropic yesterday released Claude Opus 4.8, which the company calls it “a modest but tangible improvement” over Opus 4.7. When a vendor undersells its own launch, pay attention to which number it is quietly proud of.

Amazon’s own assistant drove nearly $12 billion in incremental sales last year across 300 million users. ASA will succeed because it’s Amazon. The moat is Amazon’s extraordinary network effect; the tech is almost beside the point. Almost.

Researchers at X41 D-Sec disclosed “BadHost,” a critical vulnerability in Starlette, the open-source Python framework embedded into roughly 325 million new software builds every week. If your enterprise has stood up an AI agent in the last 18 months, some part of the stack runs on Starlette.

Environmental activist Erin Brockovich launched a website this week asking Americans to report data center concerns in their communities. She’s positioning herself as the voice against AI infrastructure expansion across the United States.

AdventHealth announced an 80% reduction in time spent on administrative tasks across its hospital system using OpenAI’s ChatGPT for Healthcare. The system operates in nine states and serves millions of patients.

Are we ready for this level of access to our personal data? Spark processes your email history, document archives, and browsing patterns to deliver contextual assistance.

Andrej joining a new pre-training research effort fascinates me because he would have no incentive to take a job he didn’t believe in.

LinkedIn's heart is in the right place, but pattern detection is a treadmill.

My Sunday essay explores a winning leadership strategy developed in the 1950s by Air Force pilot John Boyd. He had a standing bet with his students that led to the OODA loop: Observe, Orient, Decide, Act. The pilot who cycles faster wins.

The unit economics of producing software have changed. You can code from anywhere at any time on two of the best foundation model coding platforms. Actually, this is worth saying differently: everyone can code (or vibe-code) from anywhere at any time. Everyone!

Will you trust an AI to spend your money? Amazon is betting you will. Personally, I can’t wait to try it!

Cate Blanchett and a coalition of A-list actors launched RSL Media yesterday, a 501(c)(3) nonprofit built to give creators a machine-readable way to declare permissions for AI training.

Claude Platform goes further than Bedrock: managed agents, code execution, web search, and day-one access to new features through Anthropic’s native APIs. The setup takes about ten minutes. I tested it.

The other day, I asked Claude to reorganize 1,400 files across six project folders. I described what I wanted in plain English, walked away, and came back to a clean directory structure with every file renamed, sorted, and deduplicated. Think about this...

This feels like a direct response to publisher complaints about lost traffic and revenue. AI Overviews and AI Mode generate comprehensive answers without requiring clicks to source websites, which translates to real money for content creators who depend on advertising and subscription conversions.

OpenAI launched a self-serve Ads Manager for ChatGPT on Monday, open to U.S. advertisers in beta. Sam Altman once called advertising a “last resort.” I guess he’s come around.