🛎️ Eight Billion Users Now Simulated

Good Morning, AI Enthusiasts!

We used to build the internet for people. That requirement appears to have been deprecated.

RESEARCH

Eight Billion Users Now Simulated

👀 What's happening: A Harvard and MIT team released MatrAIx, a system that spins up 8.3 billion synthetic "users," one per person alive, each stitched from real-world data: census stats, social surveys, Wikipedia bios, Amazon reviews, developer polls. Point an AI like Claude or GPT at them and it plays these people out, filling surveys, browsing pages, tapping through apps, so a company can test a product on "humanity" before one real person sees it. Months of recruiting collapse into minutes.

🌍 How this hits reality: For the first time, "ask the world what it thinks" is a query you run before lunch, not a study you fund for a quarter. That's the seduction, a focus group that never says no, never costs a thing, never contradicts the boss. But the appeal is also the trap: same persona, same page, different model, and the share choosing to pay swings from 23% to 94%, according to the paper. The authors' fix is to poll several models and average the skew. You've made it cheap to ask a billion questions and left unanswered whether a single answer belongs to anyone alive.

🛎️ Key takeaway: We didn't learn to simulate humanity, we learned to hold the meeting without inviting it, and the loudest voice in the room is the AI model we paid to "fake" the rest.

CODING

The End of Cursor

👀 What's happening: Cursor, the AI coding IDE tool four MIT grads built into $4B in annual revenue and 50,000 enterprise customers, is being bought for $60 billion and quietly erased. According to The Information, the $60 billion purchase of Cursor could close as soon as this Friday, and with it the name starts to disappear and its unreleased agent shipped under a Grok label instead.

🌍 How this hits reality: Cursor won on one promise: we don't care which model you use, bring Claude, bring GPT. That neutrality is what pulled teams off Copilot. But the model layer was always going to swallow the app built on top of it, that was written the day it was born, not decided in this deal, the acquisition just collects what gravity already owed. And the real prize was never the customers, it was the data: every completion developers accept trains the very model set to replace the tool. Neutrality was the disguise on a pipeline.

🛎️ Key takeaway: Cursor's story was supposed to prove the app layer could stand alone. Its ending proves the opposite, that a company without its own model was never independent, only unacquired.

ADS

Time Is Stuffing Ads Into AI Memory

👀 What's happening: A German developer found Time serving hidden ads to AI crawlers, FAQ-shaped article versions only machines ever see. They aren't sent to live search bots but to training crawlers like ClaudeBot, which bake pages into a model's long-term memory. The pitch: why advertise to one human when you can shape the output of a whole model? Whether it even works is unclear, since model builders scrub training data and may wash the ads right out.

🌍 How this hits reality: It's easy to call this poisoning the well. But look at Time: traffic is now mostly bots, human-facing ads go unseen, revenue is collapsing. When your audience stops being people, selling to people stops paying the bills. This isn't a publisher turning malicious, it's one following its readers. The open web was built as a library for humans; it's quietly becoming feed for machines, and the businesses that lived on human attention are migrating their customers from people to crawlers, because that's where the eyeballs went.

🛎️ Key takeaway: The disturbing part isn't that Time did this. It's the structure that made advertising to a machine rational, a web where the surest way to reach a person now runs through the model reading on their behalf.

RETAIL

Retailers Still Own the Wallet

👀 What's happening: As shoppers increasingly ask ChatGPT and Gemini what to buy, retailers are racing to appear in those answers. The referrals are lucrative: Ulta reports double the purchase intent from AI, and Etsy sees shoppers discover products in chatbots. But visitors sent by AI spend 41% more per visit, so brands optimize to rank in chatbots while steering the actual purchase back to their own sites, where the customer, and the data, stays theirs. The traffic is welcome. The checkout is not for rent.

🌍 How this hits reality: The habit of asking AI first is real, but it only wins the top of the funnel, the browsing, the shortlisting. The moment money moves, people still go home. Etsy shoppers discover products in ChatGPT then return to Etsy to pay; OpenAI quietly shut down in-chat checkout and retreated to discovery. Handing over your card takes trust, not convenience, and trust is thirty years of returns, service and loyalty points that a general-purpose model simply doesn't have for now. AI got the curiosity. Retailers kept the wallet.

🛎️ Key takeaway: The habit brings people to the door, but crossing the counter still runs on something retailers spent decades building and AI can't yet borrow.

DAILY TL;DR

• GitHub added Copilot side chats and isolated worktrees so developers can run parallel AI coding sessions more safely.
• Singapore said it will help workers adapt to AI, while warning that not every job can be protected.
• X is replacing revenue sharing with Original Content Rewards to reduce incentives for AI spam and clickbait.
• Cloudflare launched Kitesurf, an agent-first browser built on Workers V8 isolates for lighter automated browsing workloads.
• Anthropic is making Claude Code Auto Mode the default for paid users, replacing most manual approvals with a risky-action classifier.
• Mayo Clinic faces a lawsuit from a former AI compliance lead who says its MAYA assistant hid error-rate and privacy problems.
• OpenAI hired former White House AI adviser Dean Ball to lead its new Strategic Futures team, bringing AI policy and governance closer in-house.

READ MORE

Let the Future Come to Your Inbox

Stay ahead without drowning in information. We turn the most important signals across AI, tech, marketing, and future products into 5-minute reads you can actually finish.

• AI Secret uncovers what really matters in AI
• Bay Area Letters decodes tech and business shifts from Silicon Valley
• Robotics Herald tracks how robots move from labs into daily life
• Marketing Secret breaks down real growth and go-to-market playbooks
• The Hardwire explores hardware, consumer tech, and what’s coming next

TOGETHER WITH US

AI Secret Media Group is the world’s #1 AI & Tech Newsletter Group, reaching over 2 million leaders across the global innovation ecosystem, from OpenAI, Anthropic, Google, and Microsoft to top AI labs, VCs, and fast-growing startups.

We've helped promote over 500 Tech Brands. Will yours be the next?

Email our co-founder Mark directly at mark@aisecret.us if the button fails.