Fascinating essay by Nikhil Suresh:

On a trip overseas, I had the privilege of a meeting with one of the Fortune 500 executives mentioned at the beginning of the post, who will remain anonymous so that they are not executed by firing squad by their board. As we were chatting, it became clear that they were very switched-on and technically competent, and they also happened to be at a company that had committed to the usual battery of exorbitant claims about their recent innovations — we’ve 100×’d our productivity, AI is the future of everything, I am but a vessel for OpenAI to make love to my wife. You know, normal things. But since I had them there without any microphones around, I asked why this was being repeated without opposition. Was it just sales fluff?

The answer was a lot more interesting. It was partially ridiculous sales material being delivered to an easily excitable audience, but this was not the dominant factor constraining honesty. Executives at their customers were saying absurd things about achieving 100× productivity, and this meant that if any executive at the vendor said that these gains were not plausible, it would undermine the credibility of the customer’s executive, be perceived as an attack (or heresy), and possibly result in an enterprise contract cancellation. And getting enterprise contracts cancelled because you wanted to opine on something that doesn’t really matter to your organisation’s mission is a great way to get fired.

The whole essay is very much worth reading, and might make you feel saner for harboring your own doubts regarding just how much has been changed in the world by generative AI. The gist of his argument is that the entire corporate world — not merely the computer/tech industry — is caught up in an AI mania that brooks no dissent. It’s a religious fervor and heretics are excommunicated. But the dissenters, who feel they must remain silent, are largely correct.

One way I’ve been thinking about AI mania is this: Computers have profoundly changed the world. There can be no dispute about that. Computers continue to improve in capability as networking effects grow and especially as Moore’s Law makes them faster and more capable at a pace unlike any other previous technical revolution. But most people are incapable of understanding how computers really work, and thus personally aren’t able to do much with them. Computers, for most people, are communication tools, not creative tools. Generative AI changes that. People with no computer aptitude are able to create things or discover things via ChatGPT and Claude that were simply beyond their ken heretofore. Per Clarke’s third law, “Any sufficiently advanced technology is indistinguishable from magic”. Generative AI feels a bit magical to all of us, even those of us with mid-to-high levels of understanding how computers work and how to make them work. But it feels like the first time ever that computers have been tools for their own creation for a lot of rather dimwitted but confident corporate manager/executive types. This moment feels like the Big Bang to them. The religiosity of this is such that there is no way to convince them that AI, as it stands today, is impressive and useful and innovative, yes, but several orders of magnitude less so than they’re imagining that it is. No way to convince them, that is, until the bubble bursts.

The European Commission on Thursday:

The Commission found that Google gives preferential treatment to its own services, including shopping, hotels, transport and sports results, over those of third parties in Google Search, thereby breaching its obligations under the DMA.

Google displays its own services more prominently in search results, including at the top of the search results page or by using enhanced visuals and filters, while similar third-party services do not have the same prominence.

So if you search for the score of a sporting event, it’s illegal in the EU for Google to show you the score, like this search I just tried. What the EC is saying is that because Google Search is so popular, it’s no longer allowed to improve. It is impossible to imagine how this benefits EU citizens.

What generates more for the EU, fines against US tech companies, or tax income from EU tech companies? Here’s a viral post on Twitter/X that claims it’s now fines against US tech companies. The amounts certainly seem to be in the same ballpark.

Julien Khaleghy, CEO of SerpApi:

The U.S. District Court for the Northern District of California granted our motion to dismiss Google’s lawsuit, marking a win not just for SerpApi, but for all who depend on an open internet. We’re pleased that the court rejected Google’s attempts to expand the DMCA to assert control over access to public pages. The internet’s founding principle — open access to usable information — is essential to driving innovation and ensuring everyone benefits from the promise of data. SerpApi will continue supporting developers, AI companies, researchers, and businesses that rely on access to public search information.

Yours truly, back in March, when SerpApi filed the motion to dismiss:

I’ve come around on SerpApi in the last few months. My initial take was that it surely must be illegal for a company to scrape Google’s search results and offer access to that data as an API. But I’ve come around to the argument that what SerpApi is doing to obtain Google search results is, well, exactly how Google scrapes the rest of the entire web to build its search index. It’s all just scraping publicly accessible web pages.

This December piece by Mike Masnick at Techdirt is what began to change my mind.

Even Google seems to understand the logical bind they’re in here, regarding scraping.

Apple Newsroom:

Apple and Ford today announced that Apple Maps will be integrated directly into Ford’s upcoming Universal Electric Vehicle (UEV) Platform through Apple’s new MapKit for Automotive SDK. Coming to Ford’s UEV Platform in 2027, this integration will deliver a beautiful and easy-to-use navigation experience powered by Apple Maps directly to the vehicle’s displays. Road-level Maps information will also enable Ford’s Latitude AI team to build a seamless hands-free driving experience.

See also: Techmeme’s roundup of commentary.

Teddy Rosenbluth, reporting for The New York Times:

On Friday, the Centers for Disease Control and Prevention announced that there have been 2,318 confirmed cases of measles so far this year. The case count toppled the record set last year, when measles infected more than 2,200 Americans and killed two unvaccinated children. More measles cases have been reported in the United States in the last two years than in all the years from 2000 through 2024 combined.

Reversing course on measles would require a nationwide push to improve vaccination rates, a prospect that experts said seemed unlikely under the leadership of Health Secretary Robert F. Kennedy Jr. He has halted funding for vaccine hesitancy research, downplayed the risks of measles and spearheaded a hunt for evidence that vaccines are unsafe.

Nilay Patel, endorsing Kamala Harris for president in 2024: “A Vote for Donald Trump Is a Vote for School Shootings and Measles”. (He missed “quagmire war with Iran”.)

“tarltontarlton” on Reddit last year: “Stupid-Americans are the New Irish-Americans, Trump is Their JFK”. I continue to think about this post frequently.

Isaac Asimov, in a column for Newsweek in 1980:

There is a cult of ignorance in the United States, and there has always been. The strain of anti-intellectualism has been a constant thread winding its way through our political and cultural life, nurtured by the false notion that democracy means that “my ignorance is just as good as your knowledge”.

I continue to think about this frequently as well.

For a long time the best native Mac IRC client was Textual, but although Textual still works, it’s been sunset by the author. Not to fear, however. Fredrik Blank just launched bIRC:

Creating an IRC client in 2026 is not something that will pay the bills : ) It would be just for the joy of programming. But the amount of work involved always made me throw the thought aside.

But one night when I had nothing better to do, I started implementing the good old spec in Swift (RFC 2812). And then I got motivated to continue, adding IRCv3 extensions and UI. Been working on it on and off on it since the end of 2025. Now I feel pretty satisfied with the result. Say hi to bIRC — an IRCv3 client for Mac.

I was never a big IRC user, personally, and haven’t used IRC in years. But I installed bIRC just because it looks so good, and took it for a brief test drive. It’s nice. And it is very Mac-assed in a modern way. It’s got a detailed documentation page too. There are (alas) many great apps that lack good documentation, but there are very few apps with good documentation that aren’t great apps. bIRC is free to download and use from the App Store, and there’s a suite of additional fun/pro features you can unlock for a one-time $13 purchase. I bought the pro upgrade despite seldom using IRC just to celebrate a killer new Mac app.

Following up on the previous item today, re: content blockers and ad-blocking, I ran into a weird situation a few weeks ago with my column “John Ternus Should Reverse Apple’s Slide Down the Advertising Slippery Slope”. That article contains this sentence, in Markdown:

And -- at this writing, still "coming soon" -- Apple is
launching [ads on Apple Maps][🗺️].

[🗺️]: https://ads.apple.com/maps

The problem is that anyone reading that article in NetNewsWire saw it rendered like this, with the words “ads on Apple Maps”¹ omitted:

It looks like I forgot to finish writing that sentence. NetNewsWire simply omits the hyperlinked words. After a few readers reported this, I immediately suspected the problem was the “ads.apple.com” domain. I’ve never linked to that domain before, and many ad blockers just block any domain that starts with a domain prefix like ads.*.

And that’s exactly what the problem is. I had no idea that NetNewsWire filtered any content whatsoever (and I sort of think it shouldn’t), but DF reader Antonio Germano searched through NetNewsWire’s open-source code and found it in the core.css file, line 49:

38  /*Block ads and junk*/
39  
40  iframe[src*="feedads" ],
41  iframe[src*="doubleclick"],
42  iframe[src*="plusone.google"] {
43      display: none !important;
44  }
45  
46  a[href*=".ads."],
47  a[href*="feedads" ],
48  a[href*="doubleclick"],
49  a[href*="//ads."],

The culprit identified, I dutifully pinged my friend Brent Simmons to report the problem. But then I had to decide what to do about it. I have no idea which feed readers are being used to read Daring Fireball’s feeds, but NetNewsWire is surely one of the most popular, and probably the most popular. It’s the feed reader I use personally. I can’t change the contents of NetNewsWire’s core.css file, so I can’t keep NetNewsWire from stripping out words that link to Apple’s ads.apple.com domain. What I could do is not link to ads.apple.com, like, say by changing my link from ads.apple.com/maps to a shorten-and-redirect URL like bit.ly/4wmXFSE that points back to Apple’s original URL.

But I didn’t do that. I don’t like playing whack-a-mole to work around bugs in other software. I have enough trouble fixing my own mistakes. I also don’t like the idea of showing everyone a “Click it and find out where it actually goes” mystery URL from Bitly rather than the direct URL. It’s a general principle thing, and also a much better policy for links that are meant to last for decades.

What I really don’t like about NetNewsWire’s ad-blocking here is that it just omits the linked text. It doesn’t remove the link but leave the words; it removes the words that are hyperlinked to the blocked domain. I’ve run into this with other content blocking extensions. For example, the Banish extension for Safari, which I recommended back in 2022. At some point a year or two ago, Banish started blocking links to Apple’s App Store domain, and it did so by removing the words that link to the App Store. That’s crazy. I linked to the App Store in my own post recommending Banish, which means if you have Banish installed and read that post, you won’t see the headline of the post (nor be able to follow the link). I think there are other content blockers that do the same thing with text that links to the App Store.²

So what do I do about that? Do I stop linking to the App Store? That’s dumb. Do I create custom URL redirects for every single link I make to the App Store? That’s work I don’t need or want to do, and adds a layer of abstraction that confuses everyone who isn’t using a content blocker with such a stupid rule. So I just ignore it. But that means an untold number of readers, who are using such content blockers, are just missing words in my articles whenever I link to the App Store. Well, so be it. Ideally, content blockers “just work” — blocking only things you want blocked, never things you don’t want blocked. Nothing is perfect however. What’s pernicious about blockers that remove the actual words that link to certain domains is that that’s not expected behavior. From the reader’s perspective, it really just looks like an editing error on the part of the website, not a content blocker with an overzealous pattern-matching rule.

Ultimately, though, when you install a content blocker extension, you take responsibility for any overzealousness in what it blocks.

---

Let me use this opportunity to make a small personal request. The display ads on Daring Fireball are unobtrusive and limited to one per page. They are also entirely private, and always have been. There not only are no cookies or JavaScript that attempt to track you, but the ads themselves are served from the daringfireball.net domain. I turn down advertisers who request to serve images (or “tracking pixels”, which are just invisible images) from their domains. (Such requests are infrequent, thankfully.) I don’t display the ads in between paragraphs mid-article, a common practice that to me is disrespectful both to the writer and reader. In short, I try to keep the ads on Daring Fireball not merely unobjectionable, but something that actually adds to the site. I actually like the ads in certain high-quality print magazines, like The New Yorker. My goal for the ads on DF is for them to be like that.

In short, I hope they’re the sort of ads readers don’t want to block. I’m pleased to say that most of the ad-blocking content blocking extensions I’ve personally tried do not block ads on Daring Fireball by default. In other cases, they do, but it’s easy to adjust the per-domain settings to allow them. In the previous post, I recommended uBlock Origin Lite (free, a bit complex) and Magic Lasso (paid, much simpler). Magic Lasso does not block ads on DF by default. I don’t remember what uBlock Origin does by default, but it’s trivial to change it to allow them.

If you use an ad blocker and it currently blocks the ads on DF, I humbly ask you to consider taking a moment to tweak the settings to permit them. (If you are using a blocker that strips the ads from DF by default, you can do me an additional favor and shoot me a short email telling me which one. I’m curious.) If you are the developer of an ad blocker, I ask you to consider allowlisting Daring Fireball by default.

If you are a reader and you really do want to block DF’s ads, please, go right ahead. I find that an odd mindset, but it’s your web browser and your call. No hard feelings. (Please further note that I have no JavaScript code that attempts to detect whether you’re blocking DF’s ads in order to nag you.)

Speaking of dickovers, here’s Cory Doctorow, writing this week at Pluralistic:

Seizing the means of computation isn’t theft, it’s bargaining. Commercial surveillance companies will tell you that by spying on you, they are simply engaged in a marketplace exchange in which you swap your privacy for access to online services. But they are running a very curious sort of market: it’s a “market” where as soon as you stop to browse someone’s wares, the stallholder gets to reach into your pocket and clean out your wallet. In “markets,” prices are announced and bargained over, not set unilaterally and extracted from anyone unwise enough to cross the threshold.

Adblocking, dickover blocking and other customizations are a way for you to bargain back, to answer the opening bid of “How about you give me all of your data forever and let me do anything I want with it?” with “How about ‘nah?’”

Doctorow’s specific advice is to switch to Firefox for your browser and Linux as your desktop OS. You’re probably not going to do either of those things. But the spirit of his advice is that you should avail yourself of bookmarklets and browser extensions that fight back against shitty ads and dickovers. There are some great options for Safari. StopTheMadness Pro is great for creating custom rules to permanently block specific page elements. uBlock Origin Lite is a free-of-charge content blocker that is incredibly effective, and very configurable (but complicated). Magic Lasso is a paid content blocker — $30/year — which includes synced blocking on iOS, MacOS, and even Apple TV. Magic Lasso offers a much more approachable configuration interface, and includes a “just turn it on” easy-to-use private VPN feature that allows for blocking ads in all apps, including Apple News.

Bargaining back is exactly what using content blockers is. There’s no reason to feel guilty about it.

(Magic Lasso has previously been a DF sponsor, but not since May 2021. I’m recommending Magic Lasso here only as a happy paying user.)

Here’s a real gem that a reader sent me. Tomtoc is a maker of laptop bags and sleeves. Visit their homepage and, of course, you’ll get a dickover asking you to join their mailing list “for exclusive updates and get a chance to win a free tech pouch every month.” But here’s the gem. If you do subscribe, and you decide later to unsubscribe, you get this:

The betting odds for the Sixers to win the NBA championship went from 20-1 to 10-1. Here’s James on Twitter/X, explaining his decision:

This is my last decision. I’m not going for money. I’m not going for family. What am I really playing for at this point?

I still want to sacrifice. I still want to work. I still want to grind. I still want to compete, to win and to have a chance at the feeling of winning another championship.

I believe I can help make the Philadelphia 76ers a championship team and I am so excited to energize a new fan base and start this incredible journey one last time.

James signed for the league minimum for a veteran: two years at just $4 million per year. He’s going to love playing here.

Quinn Nelson returns to the show to discuss OpenAI’s ChatGPT/Codex “native” app migration fiasco, Siri AI in Apple’s OS 27 betas, MacOS 27 Golden Gate, and the hottest new cell phone of the year, the Trump T1.

Sponsored by:

• Notion: Try the powerful, all-in-one Notion AI today.
• Factor: Healthy eating, made easy. Get 50% off your first box, plus free daily greens, with code talkshow50off.
• Squarespace: Save 10% off your first purchase of a website or domain using code TALKSHOW.

“From a Bond fan to Bond fans worldwide.” Nice work.

Kirk McElhearn, back in February:

I use Apple News to keep up on topics that I don’t find in sources I pay for (The Guardian and The New York Times). But there’s no way I’m going to pay the exorbitant price Apple wants for Apple News+ — £13 — because, while you get more publications, you still get ads.

And those ads have gotten worse recently. Many if not most of them look like and probably are scams. Here are a few examples from Apple News today.

One of the ads he examined was, supposedly, from a small company going out of business because the proprietor — supposedly a straight-out-of-Central-Casting kind old woman — was retiring after 26 years. McElhearn looked up the domain name and it had only just been created last year, and was registered from a company in China.

Two years ago this very week, I wrote about Apple striking a deal with Taboola to sell ads for Apple News. In that post, I quoted Eric Seufert (of Mobile Dev Memo) posting on Threads:

Regarding Taboola’s partnership with Apple: I’ve seen people claim that this is somehow hypocritical from a privacy perspective, assuming that Taboola’s somewhat obnoxious, clickbait-style ads must invasively target user profiles and browsing histories.

They don’t. They are targeted entirely contextually. That’s the point.

Want brash, garish advertising plastered all over the web? Reject ads personalization. Want relevant, informed advertising? Embrace ads personalization.

Now quoting from myself in that same article:

If you told me that the ads in Apple News have been sold by Taboola for the last few years, I’d have said, “Oh, that makes sense.” Because the ads in Apple News — at least the ones I see — already look like chumbox Taboola ads. Even worse, they’re incredibly repetitious. [...]

So while I don’t think it’s good news that Apple is partnering with Taboola, I don’t expect it to make any discernible difference in the ad quality or frequency. Maybe it will improve the variety?

Privacy is not the issue with Apple’s Taboola partnership, or Apple advertising in general. The issues are quality and user experience. I was correct that the partnership has made no discernible difference in ad quality or frequency. And it hasn’t made any improvement to variety either. Here’s a collection of ads I continue to see in Apple News — often with the same ad appearing 3–4 times in the same article, as I scroll. (I posted the same image in my recent article “John Ternus Should Reverse Apple’s Slide Down the Advertising Slippery Slope”, if you’re thinking it looks familiar.)

The ads I see in Apple News are not relevant or informed. And, to be clear, I have “Personalized Ads” turned on in Settings → Privacy & Security → Apple Advertising (which makes me wonder if turning that off might actually make the ads even worse). The ads shown in the App Store never seem contextually relevant to me. So I have no faith that the ads that Apple is supposedly on the cusp of serving in Apple Maps are going to be good.

Back in March I wrote:

I’m not going to prejudge the actual experience, and you shouldn’t either. I also do not begrudge Apple for wanting to monetize Maps. But if the addition of ads does make the Apple Maps experience worse, why won’t Apple let us buy our way out of seeing them? Netflix doesn’t force us to watch their ads. YouTube Premium is arguably the best bang-for-the-buck in the entire world of content subscriptions. Why should Apple One subscribers still see these ads in Apple Maps?

This to me is proof that Apple leadership is suffering from cognitive dissonance over their slide into advertising. They’re conflating cause and effect. The Apple brand stands for high-quality superior user experiences because the devices and software Apple makes traditionally deliver high-quality superior user experiences. The work Apple ships is what makes the brand what it is. But now they’re stricken with the mindset that it’s the Apple brand that makes something a high-quality superior experience. If it comes from Apple it must be great — that sort of thinking. The ads in Apple News and the App Store can’t be detrimental to the user experiences and at times even embarrassingly crude or downright scammy because Apple doesn’t ship things that are detrimental to the user experience or embarrassingly crude — and certainly never with even a whiff of scamminess.

My theory is that’s why Apple isn’t allowing users to buy our way out of the ads. When streaming services like Netflix let you pay more money for ad-free tiers, it’s an explicit acknowledgement that the experience with ads is worse. Apple doesn’t offer ad-free tiers for the App Store or (soon) Apple Maps¹ because, I think, they can’t bring themselves to admit that by adding ads to these services they have made them worse. If you refuse to admit that the ads make these services worse, you also can’t admit that there’s a reason to offer an ad-free experience. So the thinking is something like “Apple always focuses on the user experience, and never deliberately makes the user experience worse. Therefore, if an Apple service shows ads, those ads are high quality and don’t detract from the quality of the product or the experience of using it. Therefore, there’s no reason to offer users a paid ad-free tier, because there’s no reason to want to avoid Apple’s ads.”

It’s a textbook case of Upton Sinclair’s famous adage: “It is difficult to get a man to understand something, when his salary depends upon his not understanding it.” The obvious truth is that Apple’s ads just plain suck. They are not good at advertising-based services. And it shouldn’t be surprising that they’re not good at advertising-based services, because the incentives for advertising-based services are in direct opposition to the incentives for everything Apple actually is good at. To wit: offering superior products worth paying a premium price for.

One can argue that an ad-free Apple Maps experience ought to be included with the price of the devices Apple sells (which, of course, has been true until now). It’s even easier to argue that the App Store ought to be ad-free for everyone, based on its exclusive role in software distribution and the 15–30 percent commission Apple takes on every transaction. But if Apple simply made the App Store and Apple Maps ad-free for users paying for iCloud+ or Apple One, I’d stop complaining. That would make it clear: the full Apple experience requires not just an Apple device, but an Apple subscription. But the way things stand now, that ad-free experience isn’t even available.

Dvorak is best known for all the takes he was jacktastically wrong about, but what gave his schtick staying power is that he was technically savvy. His take here on the PS/2 was actually spot-on: that it was technically impressive in many ways, especially the way it was assembled without any cables, but that it was doomed in the market against ugly mess-of-cables commodity PCs.

“Computer Chronicles” was the real show from the 1980s that Adam Lisagor so lovingly and spectacularly spoofed with “Computer Show” a decade ago.

William Gallagher, AppleInsider:

Legendary technology columnist, pundit, author and podcast host John C. Dvorak has died, age 80. He is survived by his wife Mimi Dvorak.

Today Dvorak is best known for co-hosting the “No Agenda” podcast with Adam Curry, where he would critique mainstream news media. His death was first announced by the podcast on X, although note that Dvorak’s age is mistakenly given as 74.

From the No Agenda News newsletter yesterday:

For a man who spent decades famously predicting the catastrophic demise of almost every major technological innovation, it is with the heaviest of hearts we must announce that John’s own hardware finally gave out on Monday morning, July 20th, 2026, at the age of 80. He went peacefully, albeit suddenly, with Mimi at his side, critiquing the acting and story line of NCIS.

His death truly seems quite sudden — the most recent episode of No Agenda dropped on Sunday, the day before he died.

Chance Miller, 9to5Mac:

Flighty’s new Connection Assistant feature includes step-by-step information for your specific connection. This includes things like terminal changes, security checkpoints, passport control, and more. Connection Assistant also gives you estimates of how long each step of making a connection usually takes.

“The new Connection Assistant combines Flighty’s best-in-class flight tracking data with airport-specific procedures and statistical modeling of millions of prior flights, and then turns it all into guidance for your exact trip,” the Flighty team explained in a press release today.

I don’t know what the equivalent term to “Mac-assed” should be for iPhone apps, but whatever it is, Flighty epitomizes it. Incredibly useful for what it does, exemplary for how it looks and feels and is visually organized. It is as iOS-assed as an app can be.

The European Commission, last week:

Today, the European Commission has issued two sets of binding specification measures to Google under the Digital Markets Act.

The aim of the first specification measures is to ensure that competitors’ Artificial Intelligence (AI) services can compete with Google’s own AI services, such as Gemini, by having equal access to features on Google’s Android devices.

The aim of the second specification measures is to rebalance the playing field by giving third-party search engines access to search data that only Google Search can collect at scale.

They provide separate “Q&A” overviews of the guidance for Android AI interoperability and web search sharing, and the full guidance documents are PDFs (Case DMA.100220 for Android AI, Case DMA.100209 for web search). I suggest reading the two Q&A overviews, unless you’re having trouble falling asleep at night, in which case you’ll love the full PDF decisions.

Both decisions are interesting. With search, Google is required to share with competitors — search engines and AI chatbots alike — a massive amount of user data from Google Search user interactions. What terms people search for, what they click on in results, what languages and devices they use. It’s all ostensibly anonymized but that’s tricky when it comes to search terms. A lot of the terms people type into web search fields are to some degree personally identifying. The EC seems to be saying it’s Google’s problem to filter out things like passwords and usernames and omit them from the shared datasets. Google can charge money for this access, but only under “fair, reasonable, and non-discriminatory (FRAND)” prices, based on a Commission-defined methodology.

More interesting to me, however, is the guidance pertaining to on-device AI on Android devices. What the EC is dictating to Google is just breathtaking in scope. The EC is demanding that Google create APIs that allow third-party AI assistants to do everything Google Gemini does now, including:

• Control hardware buttons (to invoke the assistant).
• Capture anything on screen, from any app.
• Seemingly unfettered access to microphones and cameras and other sensors on the device.
• Unfettered background operation. Third-party AI assistants must be permitted to execute in the background whenever they want, for as long as they want.
• Execute their own audio models on the digital signal processor, so they can listen at all times for their own custom “Hey Dingus” wake phrases/hot words.
• Google must allow concurrent access to always-on hot word detection. So if you have Claude and ChatGPT and Grok and Meta AI installed, all of them — in addition to Gemini — must be permitted to have always-on audio detection concurrently. Google is permitted to do some vetting here, but this seems like madness.
• Third-party models get access to Google’s on-device local models.

Also, in my reading, the EC is demanding that Google make available to third-party AI assistants all information in Google’s own apps (Gmail, Google Calendar, Google Docs, Google Maps, etc.) that Gemini has access to. There is no opt-out for Google regarding data from their own apps. Nor, I think, does this guidance allow third-party apps from other developers to only support specific system-level AI models. Like, let’s say you’re Slack, and you use the APIs to make the content from within Slack available to Gemini on the device. These guidelines don’t permit Google to allow Slack to say that they trust Gemini but only Gemini. If a third-party app like Slack supports making its data available to any system-level AI provider, it must make its data available to every system-level AI provider.

There’s a lot more. Basically, though, the EC is demanding that third-party AI assistants be enabled to become part of the system software, not just apps. I’m sure some people think this is a great idea. It’s the user’s device, they should be allowed to make ChatGPT or Claude or Meta AI part of their OS if they want. It’s up to them. Put users in control.

This is how PCs have traditionally worked, but many normal people’s PCs are a mess of third-party software running in the background. That includes the Mac. If you ask a normal person “What third-party software runs in the background on your Mac or PC?” they would have no idea. It’s all just magic to them. If Google supports this guidance, it could turn Android phones in the EU into PCs. Honestly, some of the stuff the EC is requiring is lower-level than what MacOS and Windows allow third-party software to do.

What the EC’s “guidance” describes in this document is an entirely different operating system than the Android that Google has designed. The European Commission obviously thinks it is their place to design operating systems. Maybe you do too. Google obviously disagrees, and so does Apple. And so should most people who have any idea how these devices work. This is a recipe for disaster, if Google were to enact it and third-party AI assistants took advantage of it.

That second “if” is a big one. One possible scenario that I consider quite likely is that Google could spend years of engineering time and human resources building out APIs to enable all of this, in the safest and most private ways possible, and no major AI assistant adopts it. That’s the way it’s turned out with a whole slew of DMA compliance for Apple and Google. Apple built an entire complex set of APIs to enable third-party web browser rendering engines, exclusively for compliance with the DMA, and there exist no third-party web browser rendering engines for iOS. Not one. Because while the EU is a big market, it’s not big enough to justify building a custom web browser just for the EU alone.

Ways I can see this playing out, in order of likelihood (most likely first):

(A) Google enacts all of this and no major AI assistants support it because it’s only for Android, only in the EU. And it’s not like ChatGPT and Claude are seeing a lack of usage as things stand now. In this scenario Google just wastes massive time and engineering talent building APIs that never get used, and Android users in the EU get hassled with additional annoying choice and permission screens just to use the Gemini features that are built into Android.

(B) Google enacts all of this and major AI assistants do support it. Unintended results include massive privacy violations where third-party assistants exfiltrate on-device data to the cloud, Meta uses on-device third-party data for the targeting of ads, and users who take advantage of these third-party assistants see significant battery life drain as third-party assistants run without limits in the background and run expensive inference locally to save on their own server costs.

(C) Google enacts all of this, major AI assistants do support it, and there are no privacy scandals, nor any issues with battery life, because each of the companies that makes these assistants develops them with respect for user privacy and for device resources like CPU and memory consumption.

(D) Google pulls system-integrated Gemini from Android in the EU, or severely restricts its capabilities. Rather than elevate third-party assistants from apps to system software, demote Gemini to the privileges of a mere app and leave Android users in the EU without a system-integrated AI assistant.

Under all scenarios, future feature updates to system-level AI in Android will appear late or never in the EU. No future new features can debut in the EU at the same time as the rest of the world because for DMA compliance, Google will need to add support for third-party assistants to do the same things. And they’re not going to hold new features for the rest of the world waiting for that.

I’m not even sure (D) is permitted under this Commission guidance, given that Google started shipping Gemini on Android last year. The entire guidance document is written under the presumption that Google will comply by building the APIs that the guidance demands, not by achieving parity by removing Gemini system integration features. It would be awkward and unpopular for Google to ship updates to Android that remove core AI features, but that might be more palatable than the alternative.¹ Note that with iOS, to my recollection, Apple hasn’t pulled any existing features from the EU. They’ve only withheld or delayed new features. Google’s on-device Gemini horse is already out of the barn.

Lastly, although this guidance document pertains to Android, not iOS, I see no reason to think the European Commission wouldn’t demand all or most of the same things from Apple. They haven’t given Apple any guidance yet, because it’s European Commission policy to give guidance only after a DMA-designator gatekeeper ships something that is then ruled non-compliant. But it’s hard to imagine Apple accepting most of these terms. Unfettered background processing and access to the microphone, cameras, and sensors? Third-party audio models running on the hardware DSP listening for wake words? Granting third-party assistants unsupervised access to all user data from apps published using App Intents?

Apple unfortunately hasn’t shared any technical details describing its proposal for a “Trusted System Agent” that it shared with the EC last year. But whatever Apple’s vision for the Trusted System Agent is, I don’t see how the EC would deem it compliant with the DMA if they want from Apple anything close to what they are now demanding from Google. There are no checks and balances or oversight that Google is permitted to apply to third-party assistants in this guidance. If Gemini can do something, third-party agents must be able to as well. Because Gemini gets to run in the background as much as Google sees fit, third-party assistants must be permitted to run in the background as much as they see fit. And if those third-party assistant developers — OpenAI, Anthropic, xAI, and Meta — have different opinions than Google on how much CPU usage is appropriate in the background, how much RAM and storage is appropriate to consume, or how respectfully to treat users’ on-device data, well that’s just tough noogies. If the user OK’s it, then it’s OK.

A month ago, after WWDC, when this guidance pertaining to Android AI was rumored to be forthcoming, I wrote:

Google is learning the lesson Apple learned the hard way with all the existing features of iOS that were deemed noncompliant with the DMA when it went into effect. The “ship it first and ask forgiveness / hope it’s deemed compliant” strategy is not a good one in the EU.

I genuinely wonder what the European Commission thinks the purpose of Android is. Google created Android for the benefit of its own services. Google was worried about Microsoft, not Apple, at the time, but they wanted to ensure their own services were available on a major mobile platform. I’m really not seeing how it’s more attractive to Google to comply with this guidance than to just pull system-level Gemini in the EU. Either they waste a small fortune building APIs no one will use, or, they spend that small fortune buildings APIs for the benefit of their biggest competitors. I get it that that’s the intended price to pay for being a designated DMA gatekeeper. But what’s the motivation for Google to do this rather than just walk away from system-integrated AI on Android in the EU? It certainly doesn’t look like the competing platform, iOS, is going to offer it in the EU anytime soon either.

Debugging SSO, managing users, adjusting auth policies, configuring branding: every configuration task has lived behind a UI that only a human can drive.

The WorkOS MCP server gives agents the same access as your dashboard login. Hundreds of operations, discoverable at runtime. Connect in one command via OAuth, with scoped tokens instead of a master API key. Pass a screenshot of your marketing site and ask your agent to match the login page. If a human had to do it before, an agent can do it now.

Connect your agent →

Ben Thompson, on the hype regarding Kimi K3, at Stratechery:

This is a point that bears repeating: because U.S. open weight model makers must follow the frontier labs’ terms of service, they (1) are worse than Chinese alternatives and (2) end up distilling the distillation, just with a detour through Chinese labs. Wouldn’t it be better if western open weight model makers could go to the source?

To that end, here’s an even more interesting question around distillation: why exactly is it bad? After all, what are large language models but the distillation of all of the knowledge on the open Internet, scraped by the frontier labs and distilled into the models that are themselves being distilled? Who is exactly being wronged here?

In fact, this paradox is the solution. I believe that open weight models are good for innovation (and, per the above, I think that labs on the frontier will be fine), but it’s a problem to be dependent on China. The U.S. should pass a law that (1) makes explicit that collecting data for training models is fair use, and (2) bars terms of service that forbid distillation, for U.S. companies at a minimum. Stopping distillation — which is literally just querying the API — is nearly impossible; the U.S. should go the other way and lean into a new copyright policy that both indemnifies the labs and also guarantees that what they learned fuels further innovation for everyone else.

To be clear — and Thompson emphasizes this point too — the leading Chinese models like K3 aren’t good only because of distillation. They’re not entirely rip-offs. But distillation is clearly an essential part of the formula they’re using to keep releasing models that are 6–9 months behind the U.S. frontier state of the art. So the Chinese treat all models as “open”, regardless of the terms of service. But anyone in the U.S. or other western countries that respect copyright — or contracted terms of service — who wants to distill a good model has to wait for the Chinese to release their models, like K3.

And that second paragraph I quote above distills (sorry) exactly why I want to bring out the world’s smallest violin to play a sad song for OpenAI and Anthropic regarding their objections to their frontier models being distilled against their terms of service.

My thanks to Paper for sponsoring last week at DF. Paper is a new professional design tool where every layer is real HTML and CSS. Your design is already code, which means fewer handoffs, translations, and layers of abstraction between what you design and what you actually ship.

Paper works both ways: code to design, and design to code. Connect any AI agent through MCP and the agent can read and edit your designs. Or use Paper Snapshot to capture your live site as editable layers and make new iterations. Agents handle the tedious work so you can focus on your craft and the decisions that matter.

Try Paper at paper.design (great domain name), and check out the documentation and FAQ for more info.