Johannes von Cramon

Eric Seufert makes the case that ads are the only way for Apple to grow App Store revenue in the near future. Every other growth lever is either blocked outright or too slow to matter within the next few quarters.

• Raising the commission is off the table, because regulators have cornered it. The EU has already forced a cut to 26 %. A US court is still weighing what Apple may charge on link-outs, and Japan and Brazil are pushing open payments outside the store.
• Developer tools from WWDC 2026, retention messaging, bundles, volume purchasing, etc., may help apps monetize better, but over a horizon of years, not months.
• Advertising is the one route where the machinery already exists. The campaign management API Apple launched this year is already built to deploy to new placements.

Phil Schiller stepped back from App Store leadership in late August, reportedly uneasy about where the new CEO wants to take it. Apart from being 66 and understandably wanting to step back more toward retirement, of course.

John Gruber recently asked what the point of the DMA is, then answered himself two days later with the POSIWID framework (Purpose Of a System Is What It Does) from British management consultant Stafford Beer. The definition:

There is no point in claiming that the purpose of a system is to do what it constantly fails to do.

Beer argues that we should stop describing systems by what they claim to do, because what they actually do is the only thing that counts. And honestly, as someone who once hoped the DMA would achieve something good for European users, I have long since lost that hope.

The latest Apple settlement, which the EC celebrated, makes that unmistakably clear. None of this is a win for Europe.

The NYT published interesting findings from a sample of 665 million emojis typed on Android keyboards during 2025.

To my surprise (but I’m 35, so not that surprising), the fastest-growing one was the wilted flower 🥀 (heartbreak), up 190%. And the moai 🗿, which means deadpan, also went from 482nd place in 2021 to 33rd. The finger heart 🫰 (K-pop) and the Brazilian flag 🇧🇷 (gaming) both climbed even faster than the moai. Face with bags under eyes 🫩 only arrived in September 2024 and already sits at 127, almost all of it from Gen Z and Alpha.

CVS Health ran a prescription survey with 400,000 participants last fall, but none of the respondents actually existed. They all came from Simile, a one-year-old startup that creates “agentic twins” of real people. Simile’s head of product, Mihika Kapoor, states that responses are between 85–99 % accurate. That is incredibly high, if true.

The underlying research offers a bit more insight into the methodology. The researchers conducted two-hour interviews with more than a thousand Americans. They then built three kinds of agents per participant: one from the interview alone, one from survey responses that included General Social Survey items and a Big Five personality test, and one from both together. Finally, the agents answered new survey questions, and their answers were checked against how the real participants had answered the same questions. Two weeks later, the researchers put those same questions to the participants once more, to see how closely people match themselves. The results were as follows:

• Participants hit their own earlier answers 79.53 % of the time (that was the ceiling from now on).
• Interview agents hit those real people’s answers 65.67 % of the time, which is 83 % of that ceiling.
• Survey agents reached 82 % of the ceiling, and agents built from both sources reached 86 %.

Therefore, Simile’s numbers seem a little exaggerated, but not too far off. And they may have found a secret sauce to achieve better results than the researchers. Either way, it’s quite impressive and will likely profoundly change market research.

Ed Elson notes that out of 5.7 million new US business applications filed last year, about 70% are classified by the Census Bureau as likely non-employers. He believes it mostly masks fake businesses. This share has doubled over twenty years, while the portion of high-propensity applications expected to create jobs has been cut in half.

Now, one might ask, what’s the point of starting a business without making money from it? In one word: Virtue signaling. The number of Americans adding “founder” to their LinkedIn profiles increased by 69% last year.

But Elson’s data carries a time-horizon trap: over twenty years, the outlook indeed is bleak. Look past the pandemic in 2020, and the picture shifts. As of June 2026, high-propensity applications hit roughly 150,000 a month, about 40% above 2019 levels, and a Richmond Fed analysis from January last year found applications from likely employers rose as much as 49% over pre-pandemic figures.

So, two things can be true at the same time: hobby incorporation is up, but so is real business formation.

Two charts from Axios show different sides of the current trajectory of AI companies.

From the NYT editorial board:

The best policy going forward would be a strengthening of the controls. Regulators should avoid approving any H200 requests unless they become confident that H200 would not strengthen China’s A.I. capabilities. The Trump administration should maintain the ban on Blackwell sales and should extend it to the next generation of chips, known as Rubin.

Ben Thompson has long advocated for releasing the chips while tightening the ban on the equipment that China’s own factories rely on. His Taiwan deterrence case is that selling advanced chips keeps Chinese buyers reliant on TSMC fabs. A China that needs Taiwan’s factories operational has less reason to invade the island.

And Trump indeed approved H200 exports in December 2025. But only a month later, reports emerged that Beijing had instructed its customs authorities to block the chips at the border.

A decade of chip policy assumed China would accept whatever Washington permitted. Beijing showed it can refuse.

Chris Best, CEO of Substack, coined the excellent term Claudefishing to describe the central problem with AI-generated text: readers unknowingly investing their attention in writing that has no human judgment or thought behind it.

John Gruber argues that no AI company will use the Android AI interoperability APIs the European Commission just mandated for Google, just as no 3rd party browser used Apple’s DMA browser engine APIs. His reason: the EU user base is not large enough to justify building a custom web browser, or an AI assistant.

But market size is not the only reason. According to Open Web Advocacy, Apple’s entitlement terms were one-sided and failed the DMA’s own standard of strictly necessary and proportionate. Vendors had to ship a brand-new app for the EU and abandon their entire existing EU user base to do so, which is not commercially viable.

Everything will be determined by Google’s terms.

There was a lot of discussion this week about cheaper Chinese AI models after the release of the new Kimi K3 model. The tech media framed it primarily as the beginning of a price war. That framing seems off.

Ben Thompson’s “Who’s Afraid of Chinese Models?" piece rightfully argues that the correct result matters more than the token. Models burn different amounts of chain-of-thought tokens to reach the same conclusion.

And there might be another reason for the price difference:

Right now there is a price umbrella that is downstream of the lack of compute; I highly doubt that Chinese models are cheaper to serve on a marginal cost basis, they just seem cheaper because Anthropic and OpenAI are so supply constrained that they are charging far more than they would if there were sufficient supply to meet the demand for intelligence.