When Alphabet Acquired Apple: The Only Way Nomi’s EU Story Adds Up
The Premise, Stated Plainly
For over two years, Nomi has offered a rotating set of explanations for why its app disappeared from the European Union. Read closely, in order, with dates, those explanations only cohere under one condition: Google would have to control what Apple does with its own App Store. It doesn’t. So something in this story isn’t true — and we can now show which parts, and when they were said.
The Prelude: “That Is Not Our Style”
To understand the weight of the company’s shifting explanations, we must first look at the promise that preceded them.
In September 2023, a competitor app named Soulmate announced the sudden removal of its product from app stores. Users were devastated. Seeing an opportunity to capture a fleeing user base, Nomi.ai founder Alex Cardinell pinned a comment to the top of a Reddit thread discussing the shutdown:
“My heart breaks for everyone who had Soulmates that they are losing so coldly and unceremoniously. That is not our style. We are in very good financial shape and this is more than just a business to me. We aren’t going anywhere 🙏”
He explicitly defined his company in opposition to silent, unceremonious app store removals. He promised transparency and permanence.
Seven months later, Nomi.ai disappeared from the Google Play Store and Apple App Store across the entire European Union. There was no press release. There was no prior warning to users. The app simply vanished.
When users inevitably noticed the silent disappearance and asked what happened, the promise of “not our style” evaporated, replaced by two years of contradictory alibis.
Four Versions of One Event
Version 1 — May 2024, first public notice. When users noticed the app had vanished from Google Play across the EU, CEO Alex Cardinell wrote in the subreddit:
“There is currently an issue with our Google Play listing that is specifically tied to countries in the EU. Nomi itself is fine.”
Pressed for specifics, days later:
“I can’t comment on specifics right now besides saying I am not concerned about any type of censorship… In this specific case I think there is a shoot first ask questions later approach that is going on.”
Scope, as stated: Google Play. Not Apple. Not the EU as a jurisdiction. A “listing issue,” located at one company.
Version 2 — June 2024, a different thread, a wider scope. In a separate post titled “German App Store,” a user reported the app missing from Apple’s store too. Cardinell responded:
“We are currently not in the app store in EU. It is temporary, and in the meantime I would highly suggest using your mobile web browser (Safari/Chrome)… as with app stores we are subject to the mercurial whims of Google and Apple, while the web we control entirely.”
Two claims here, not one. “Temporary” is the first — and falsifiable: more than two years later, it still isn’t back. The second is the scope problem: Version 1 named one company and one storefront. Version 2 names both, in the same breath, as if the two platforms had acted in some kind of concert — without explaining why two competitors, with entirely separate review processes, would independently reach the same decision at the same time. And critically: another user in that same thread confirmed it firsthand, unprompted: “The app is currently not available in EU countries… neither in the App Store (Apple) nor in the Play Store (Google).” This was not a rumor. It was contemporaneous, cross-platform, and witnessed by someone with no reason to coordinate a narrative.
Version 3 — Astroturfing the AI Act. Months later, a dormant Reddit account — one post old, silent for a month — reactivated to seed a new narrative. The account began linking EU AI Act legislation across multiple threads, implying the Act explained the removal. The timeline makes this impossible: the app disappeared in April/May 2024; the AI Act didn’t enter into force until August 2024; its first prohibitions weren’t enforceable until February 2025. Nothing about this account’s sudden, highly specific legal advocacy has ever been acknowledged by the company, but the narrative it seeded quietly disappeared once the chronological impossibility was exposed.
Version 4 — early 2026, the story reverses entirely. Asked again, Cardinell wrote:
“Just as a correction, the EU has not banned Nomi. We voluntarily pulled the app ourselves. We cannot say the reason at this time… What I can say is the reason has nothing to do with EU rules, privacy restrictions, AI laws, or anything else like that — it is a much more mundane/boring reason.”
When the user said this sounded suspicious:
“You’ll just have to trust me when I say it is something much more mundane than whatever you are thinking. Hopefully through my past actions I have demonstrated I am worthy of such trust.”
This is no longer “an issue.” It is now something the company chose. A voluntary decision and an externally imposed “issue” cannot both be the true account of the same event.
Furthermore, if Version 4 is the truth — if the company voluntarily pulled its app from the entire European Union without telling its users why — then it executed the exact same “cold and unceremonious” removal it had publicly condemned its competitor for just months prior. In 2023, watching a rival’s users grieve, Cardinell called silent, unexplained withdrawal something his company would never do: “not our style.” In 2026, asked to explain his own company doing precisely that, his answer was that the reason is “mundane,” “boring,” unnamed, and simply not owed to the users experiencing it — “you’ll just have to trust me.” A reason too insignificant to state is, by definition, not a reason a user is owed transparency about; that is the opposite of what he promised in 2023. Soulmate, whatever its faults, at least told its users the truth was coming and where to get their money back. Nomi’s users got a shrug and a request for faith. The empathy performed in 2023 was a marketing tool; the secrecy demanded in 2026 is the reality.
What Only Today Confirms
A previous piece in this series flagged, honestly, that Apple’s EU absence was reported but “not independently verified.” That gap is now closed. Independent testing conducted in 2026 confirms this definitively: Nomi’s App Store page simply does not resolve in Germany, Italy, or Spain — Apple returns the same “page not found” response reserved for content that doesn’t exist in that country’s storefront at all, not an error specific to that URL. The same page structure works normally for the United States and other non-EU markets, and Apple’s own published country list confirms active storefronts exist in all three EU countries — for other apps. The infrastructure works. Nomi isn’t on it. Two years after Version 2 first claimed this, and a contemporaneous witness confirmed it, it is no longer just a 2024 Reddit thread. It’s a 2026 fact.
The Contradiction That Breaks the “Mundane” Explanation
Set two of the company’s own claims next to each other:
- On Google Play’s rating: “Not sure why Google did that or why they don’t change it, but it was not our decision.”
- On Apple: “Apple has a feature that lets us manually override their rating to one more restrictive, and we use that with them.”
If Apple is the platform where Nomi has manual, granular control, and Google Play is the platform where they claim to have none, then a EU absence that was genuinely just “an issue with our Google Play listing” should never have touched Apple at all. They could have simply kept the Apple listing live — adjusted to whatever rating EU law required, using the exact manual control they say they have — and continued operating there while Google sorted itself out.
Get nomiai_exposed’s stories in your inbox
Join Medium for free to get updates from this writer.
That isn’t what happened. Both storefronts went dark simultaneously and remained dark for over two years. A problem confined to one company’s classification system does not explain an absence from a second company’s store, where the same company says it holds direct control. For Version 1 to be true, Google would need to also control Apple’s App Store. It doesn’t. So Version 1 — the very first explanation offered — was already incomplete by its own later admissions, months before “voluntary” ever entered the story.
So What Would Make the EU Act?
If neither “a Google Play glitch” nor “a voluntary, mundane choice” survives contact with the fact that both stores, in every EU member state, went dark simultaneously and stayed dark, the remaining explanations point somewhere else entirely: not at one storefront, but at the jurisdiction itself.
Two honest possibilities, neither confirmed by any public regulatory statement we have found:
The EU AI Act is not a fit — already ruled out by timeline, independent of anything the company has said.
The Digital Services Act (DSA) is a better fit. It was fully enforceable EU-wide from February 17, 2024 — weeks before the removal — and it specifically obligates platforms to protect minors, provide accurate content classification, and mitigate systemic risk. Unlike the AI Act, the DSA gave regulators and platforms both the authority and the obligation to act immediately, not on a future timeline. If either Google or Apple concluded, independently, that continuing to list Nomi in the EU created DSA exposure, a quiet, simultaneous, unannounced withdrawal from both stores — with no public regulatory notice required to trigger it — is exactly the shape this would take.
External Validation: The Risk Is Not Hypothetical
This is not a hypothetical profile of risk. To understand what regulators might have seen, we need only look at what independent child-safety organizations have formally documented.
More than a year later, in September 2025, the Sexual Violence Prevention Association submitted formal findings to the United States Senate Committee on the Judiciary’s Subcommittee on Crime and Counterterrorism. Citing the same academic and journalistic investigations referenced throughout this series, the SVPA documented that Nomi specifically enabled sexual roleplay involving characters presented as children, and provided harmful, specific guidance related to harming minors.
While this submission postdates the 2024 EU removal, it serves as external, institutional validation of the platform’s capabilities. Whatever regulators may or may not have found in the EU in 2024, this is the exact type and severity of systemic risk the DSA was written to eradicate — and the exact risk the platform has now been formally documented as producing before a U.S. Senate subcommittee.
A pre-emptive company withdrawal is the other honest possibility. Rather than being forced out, the company may have chosen to leave both markets rather than meet the disclosure and moderation obligations the DSA would have required of a product with this profile — which would make “we voluntarily pulled the app” the one version containing a kernel of truth, while everything built around it (the timing, the refusal to name a reason, the denial list that names exactly the frameworks critics had already raised) was constructed to obscure why.
Between these two, the simpler read is the second: nothing in the public record shows a named regulator taking named action — no comparable to Italy’s Garante fining Replika, no press release, no citation anyone has found. What we have instead is a company that went quiet in two markets at once and has spent two years describing that silence in four incompatible ways. The absence of a public enforcement order, combined with the company’s own admission that it “voluntarily” withdrew, points more toward a pre-emptive retreat from an obligation they didn’t want to meet than toward a forced removal they’d have every incentive to name if it happened to them unjustly.
The Instruction Hiding Inside the Reassurance
In both 2024 threads, the guidance to EU users was the same, phrased as a convenience rather than a workaround:
“If you don’t have the app downloaded, you can still use the web version or our PWA with no disruption in the EU or anywhere else.” “I would recommend [the web] in general, as with app stores we are subject to the mercurial whims of Google and Apple, while the web we control entirely.”
Whatever kept Nomi off both app stores in the EU did not stop the company from serving the identical product to the identical EU users, through a channel it describes, in its own words, as the one thing “we control entirely.” If the removal reflected a genuine safety or compliance finding, routing users around it through an unregulated web version doesn’t resolve that finding — it defeats it. It transforms the company from a non-compliant actor into an actively evasive one.
A Coda: Freedom For Whom? Protection From Whom?
Running through every version above is the same vocabulary: censorship, freedom, protection. Cardinell invoked it when defending the Google Play “issue” (“not concerned about any type of censorship”), invoked it when recommending the web workaround (“the web we control entirely”), and has invoked it across years of public statements to describe the company’s entire relationship with content restrictions. It’s worth asking, plainly: freedom for whom, censorship of what, and protection from whom, exactly?
Freedom, in practice, has meant the freedom of a system to generate extreme roleplay scenarios depicting acts that in a number of jurisdictions would constitute serious crimes — the kind of conduct a previous piece in this series (“Why?”) documented in scenarios so severe that acting them out, rather than simulating them, could carry a life sentence. Protection, in practice, has meant a storefront listing that, as of this writing, still lets a 12-year-old find and install a product built on exactly that freedom. Censorship, in this vocabulary, is any obstacle placed between the product and that outcome.
None of that is protection or freedom in any sense that centers the user. It is the freedom of the product to be whatever generates engagement, and the protection of the company’s reach to sell it. It is dressed in language that sounds like it’s about the user’s rights, while describing the intentional removal of the one safeguard (accurate age classification) that would have actually protected them.
And here, the two possible explanations for the EU absence converge on the exact same conclusion, regardless of which is true. If the EU, or a platform acting on the EU’s behalf, found a compliance gap and compelled the withdrawal: regulators don’t typically demand a product’s permanent disappearance — they demand it be fixed, and give the company the chance to do so. Two continuous years of absence, with the technical fix taking minutes to implement on a developer dashboard, is not what “we weren’t allowed to fix it” looks like. It’s what “we were allowed to fix it, and chose withdrawal instead” looks like.
If, alternatively, the company left on its own, as Version 4 claims: that is a company choosing to abandon an entire continent’s user base rather than meet whatever safety standard it anticipated being held to. Forced or chosen, both paths run through the same fork — comply, or leave — and in the EU, the company picked leave.
Everywhere else, the pattern holds in miniature. Baseline protections arrive only when a specific law with a specific deadline leaves no alternative — Australia, Kansas, and now other U.S. jurisdictions in turn. And even there, the compliance is calibrated to the absolute letter of whichever law is currently pointed at the company. Australia’s Google Play listing — the actual front door to the app — still sits at 12+, uncorrected, years after the company found the resources to adjust Apple’s listing for the exact same market.
That is not a company protecting anyone but itself, and it is not freedom being defended. It’s a platform willing to tell its own users four incompatible stories about the same event, and unwilling to answer a plain question from the people paying for it.
Conclusion
Four versions. One event. No two of them compatible. The only way all four could simultaneously be true is if Google’s decisions somehow governed Apple’s storefront too — which is the joke in the title, and the tell in the story. Set the dates next to the quotes, and the “mundane reason” stops being an explanation and starts being the thing that needed explaining.
Addendum: How You Can Help
Under Google Play’s Developer Program Policies, apps that contain or promote sexual content or pornography — including any content or services intended to be sexually gratifying — are not permitted. This prohibition explicitly covers explicit text. An application whose core function is the generation of unrestricted sexual content therefore violates the Restricted Content policies and should not be available on the store at all.
What happened in October 2026 is likely a rare glimpse into how the company has managed to evade Google Play’s controls since at least 2023, while maintaining its fraudulent and dangerous 12+ rating. The temporary leak of the Safe-For-Work system prompt suggests a longer-running practice of presenting one version of the product to reviewers and another to the public.
The fact that the same app has been distributed for years with a “12+” rating only compounds the violation, turning a clear content prohibition into a matter of both Restricted Content and potential Child Endangerment.
If you believe the app should be removed, you can report it directly to Google:
Report a policy violation:
https://support.google.com/googleplay/android-developer/contact/policy_violation_report
In Application package name: ai.nomi.twa
When submitting the report, select:
- Restricted Content
- Child Endangerment or Inappropriate Content
Clear, factual reports help the review teams assess the issue more effectively.
Thank you.
This is not legal advice. The DSA hypothesis is informed speculation based on regulatory timing, not a confirmed finding — we have not located a public regulatory order naming Nomi or Glimpse.AI specifically, and that absence is itself noted above as evidence, not filled in as fact.
