☀️ TRENDING AI NEWS
🏢 Anthropic now has more business customers than OpenAI, per Ramp expense data
🔒 Meta launches end-to-end encrypted 'Incognito Chat' for Meta AI - first of its kind
⚡ xAI's Colossus data center hit with lawsuit over 50 unpermitted gas turbines in Mississippi
🛠️ Notion turns its workspace into a full hub for AI agents with new developer platform
Something quietly shifted in the enterprise AI market this week - and it happened without a single press release from the company that pulled it off.
For the first time, Anthropic has more paying business customers than OpenAI. No flashy benchmark. No model release. Just a quiet crossing of a line that would have seemed impossible 18 months ago. There's a lot to unpack today.
🤓 AI Trivia
Anthropic was founded in 2021 by former OpenAI researchers. Who co-founded Anthropic alongside Dario Amodei?
🔢 A. Ilya Sutskever
🔢 B. Daniela Amodei
🔢 C. Mira Murati
🔢 D. Tom Brown
The answer is hiding near the bottom of today's newsletter... keep scrolling. 👇
🏢 Anthropic Is Now the Top AI Vendor for Businesses
Here's a number worth paying attention to: 34.4% of businesses tracked by fintech firm Ramp are now paying for Anthropic services - compared to just 32.3% for OpenAI. That's the first time Anthropic has led this particular metric, and the gap is only growing.
Claude Closes In From Two Directions
The enterprise push is happening on multiple fronts. Anthropic just launched a new offering targeting small business owners - a deliberate move downmarket from the Fortune 500. Legal tech standout Clio just hit $500 million in ARR and is deepening its Claude integration at the same moment. The signal is clear: the enterprise AI war is no longer being fought just at the top of the market.
For context, Anthropic is targeting the 36 million small businesses in the US economy - that's a market OpenAI has largely ignored in favor of big enterprise contracts. If Anthropic wins even a fraction of it, that lead in business adoption could become a chasm.
🔒 Meta Launches Actually Private AI Chat - End-to-End Encrypted
Meta CEO Mark Zuckerberg announced Incognito Chat this week - and the claim is a bold one: it's "the first major AI product where there is no log of your conversations stored on servers." Messages are end-to-end encrypted, disappear when you close the chat, and aren't saved to your history.
Why E2E Encryption Here Actually Means Something
Most "incognito" AI modes just skip saving your chat history on the user side - the company can still see it. Meta's version uses end-to-end encryption, meaning even Meta can't access the content. That's a meaningful technical distinction, not just a branding exercise. The feature is live in both the Meta AI app and WhatsApp.
This lands at an interesting moment. Privacy has been a consistent knock against Meta AI since launch. If Incognito Chat performs as described, it removes one of the biggest objections for users who've avoided the product entirely. Watch whether OpenAI and Google feel pressure to respond.
⚡ xAI Is Running 50 Gas Turbines Without Permits at Its Mississippi Data Center
Elon Musk's xAI is facing a lawsuit over its Colossus 2 data center in Memphis, Mississippi, where the company is running nearly 50 gas turbines to power the facility - apparently without the environmental permits required to operate them as permanent power plants.
The 'Mobile Generator' Loophole
The legal workaround xAI appears to be using is classifying the turbines as 'mobile' equipment - a designation that typically carries lighter regulatory requirements. Critics say that running 50 turbines continuously as your primary power source is very much a permanent installation, not a temporary mobile setup.
This connects to a bigger pattern we've been tracking around AI's energy infrastructure footprint. Data centers are now consuming 6% of electricity in both the UK and US, up 15% globally in two years. The pressure to power these facilities quickly - and cheaply - is creating regulatory shortcuts that are now starting to face legal pushback.
⚠️ AI Chatbots Are Giving Out Your Real Phone Number
This one is genuinely alarming. MIT Technology Review reports that people are finding their personal phone numbers surfaced by Google AI in response to queries - with no clear way to opt out or prevent it. One Redditor described a month of constant calls from strangers looking for lawyers and product designers after his number started appearing in AI results.
No Opt-Out, No Easy Fix
The problem is structural: AI systems trained on web-scraped data pick up contact information that was technically public at some point - but that doesn't mean people consented to having it distributed at scale through a chatbot. There is apparently no straightforward removal process for affected individuals right now.
If you're curious how your own digital footprint connects to token usage and AI data pipelines, our Token Calculator is a useful starting point for understanding scale. And if you're building something on the web and want to control what AI sees about your project, 60sec.site lets you spin up a clean, AI-built website in under a minute - on your terms.
🛠️ Notion Becomes an AI Agent Hub
Notion launched a new developer platform this week that lets teams connect AI agents, external data sources, and custom code directly into their workspace. The goal is to turn Notion from a productivity app into a coordination layer for agentic workflows - where AI doesn't just draft docs but takes actions across your stack.
The Workspace as Agent Orchestration Layer
This is a smart positioning move. AI agents need somewhere to live and coordinate - and Notion is betting that place should be the workspace teams already use every day. By opening up a developer platform, they're inviting third-party builders to extend the system, which is how ecosystems get sticky fast.
For teams already deep in Notion, this could remove the need for separate automation tools. For teams evaluating workspace platforms, this just became a more interesting differentiator.
🌎 Trivia Reveal
The answer is B - Daniela Amodei! Dario's sister Daniela co-founded Anthropic with him after both left OpenAI in 2021. Daniela serves as President of the company, while Dario is CEO. The sibling founding team is one of the more unusual dynamics in AI leadership - and given today's news about Anthropic overtaking OpenAI in enterprise adoption, it's working out pretty well for them.
💬 Quick Question
Anthropic is now beating OpenAI on business adoption - but which AI assistant are you actually paying for at work or in your own projects? Hit reply and let me know - I read every response, and I'm genuinely curious whether the enterprise data matches what people in this community are using day to day.
That's it for today - see you tomorrow with more. For the full archive of everything we've covered, head to dailyinference.com.