The dating app is the most dangerous social product ever built. Not because it is malicious, but because it is honest about what it is.

Open architecture. Partial visibility. Anyone can join, anyone can browse, and no one can hide behind a plausible deniability that the platform itself does not already grant them. The predator does not need to lurk in the shadows of a legacy social network, building a profile for months before striking. He needs only to swipe.

The dating app aggregates him. It builds a single, searchable, frictionless pool in which every man who wants access to women has a reason to be present, and it removes every social mechanism that would otherwise filter him out. The village had gossips. The market has none. Every man with an incentive to hunt has an incentive to be on the app, and the app has an incentive to keep him there, because his engagement is indistinguishable from anyone else’s.

This is why the concentration is so extreme. The Australian Institute of Criminology found that almost three in four people experienced online sexual harassment, aggression or violence from someone they met on a dating app, while one in five reported being threatened. Research from Brigham Young University analyzing over 3,000 sexual assault cases found that dating-app-facilitated sexual assaults happen faster, and are more violent, than when the perpetrator meets the victim elsewhere. The concentration is not an accident of scale. It is a selection effect, and it runs in the direction the platforms refuse to name.

The Historical Pattern: Relationships Form in Contexts, Not in Markets

Across cultures and eras, most relationships have formed in contexts not directly related to dating. The village, the church, the workplace, the guild, the extended family network, the neighborhood, the club, the cause. People met through shared activity, shared obligation, and shared community. The relationship emerged from the context rather than being the purpose of it.

This is not nostalgia. It is a structural observation about how human mating works. The contexts that produced relationships were contexts in which people were doing something else — worshipping, working, organizing, building, competing, creating. The relationship was a byproduct of proximity, repeated interaction, and shared purpose. The context filtered for compatibility before anyone was consciously evaluating compatibility. The community provided the accountability that made the relationship durable.

The moment you make an app about dating, you invert the structure. The context is no longer the filter. The app is the context, and the app’s purpose is the relationship itself. This attracts a specific segment of the demographic: the people who are explicitly looking, explicitly evaluating, and explicitly optimizing for the outcome. That segment includes the honest seeker. It also includes the predator, because the predator is the person who most benefits from a context in which everyone is looking for a relationship and no one is embedded in a community that would hold him accountable.

Identity-gated dating apps serve the segment that wants to date explicitly. They are a legitimate product for a legitimate market. But they are not the solution for the mainstream. The mainstream will eventually gravitate back to the communities that resonate with their personal interests, using the already available identity infrastructure that the platforms offer in their local area, eIDAS identity in the EUDI Wallet for Europeans, Digital ID in Kenya, or UIDAI/Aadhaar in India, or whichever is available for you by your issuing government service. The public profile becomes the portable credential. The community becomes the context. The relationship emerges from the context, as it always did.

The Legacy Platform: The Architecture of Deniability

On Instagram, Facebook, LinkedIn, Snapchat, and TikTok, the predator and the fraud hide behind plausible deniability. The profile is public, but the behavior is private. The messaging is encrypted or ephemeral. The evidence is difficult to collect. The platform’s architecture is designed for broadcast, not for intimate interaction, which means the mechanisms for detecting predatory behavior are weak and the mechanisms for reporting it are diffuse.

The exception, partially, is LinkedIn. Its partial traceability to real people — work history, education, colleagues — creates a friction that the other platforms lack. But the friction is not a wall. LinkedIn is currently adding verification features aimed at making fake professional identities, invented work histories, and company impersonation harder to pull off, including colleague vouching, employer controls over false affiliations, and workplace verification requirements. The platform’s own framing is that authenticity is becoming the most valuable currency on the internet. But the friction is also being eroded by AI. Generative AI has reduced the cost of creating convincing headshots, biographies, résumés, and recommendations, which is precisely why LinkedIn is introducing the features now rather than later.

LinkedIn’s identity verification system, likewise platform-specific identity verifications on other U.S. based social media, is also problematic simply because the platform comes from the U.S. Genuine professionals and industry representatives from outside of the U.S. are structurally disincentivized from providing their credentials onto a platform subject to the U.S. Cloud Act and letting this platform map their professional and social graphs and index and analyze their private conversations. Most opt for partial exposure in lack of local alternatives. This is simply a fact that cannot be changed, whichever your position on the reliability of the U.S administration or its policies may be.

The broader picture is worse. Romance fraud is up 139% in one year, and dating apps make up 35% of romance scam claims, while 44% begin on social media. In 2023, 22% of people in the UK were victims of being catfished by a false digital identity. In the fourth quarter of 2023, Facebook removed nearly 700 million fake accounts — after removing 827 million in the previous quarter. The volume is industrial.

The Static Profile: A Solution That Becomes the Problem

Tying the account to a real identity is the obvious fix, and the platforms are doing it. Tinder’s Face Check requires new users to submit a video selfie that creates a 3D scan of their face and verifies its similarity to their profile pictures. Tinder reports a 60% reduction in exposure to potential bad actors and a 40% decline in bad actor reports. Bumble has made photo verification mandatory for new users in the US and offers government ID verification combining facial recognition with human review. The Australian industry code of conduct requires platforms to detect potential incidents of online-enabled harm, terminate the accounts of some offenders, and show users how well platforms are meeting their obligations.

But identity verification solves only half the problem. The static public profile — the verified, unchangeable, searchable record of who you are — introduces a new set of failures. The first is saturation. The social graph hits a ceiling. Once you have connected with everyone in your relevant network, the profile becomes a known quantity. The verified badge stops signaling anything new. The user who was once a novel connection becomes a familiar one, and the platform’s utility for finding new connections declines.

The second failure is the pivot to branding. When the social graph saturates, the user does not stop using the platform. The user pivots. They become a brand. The profile is no longer a representation of a person seeking connection; it is a marketing asset. The user gains followers for their funnel rather than social connections. The content becomes optimized for the algorithm, not for the human on the other end. The verified identity is now a verified brand, and the brand is the problem. The user is no longer a person to be met; they are a commodity to be consumed.

This is the structural failure that the identity verification discourse refuses to see. The static profile solves the problem of authenticity. It does not solve the problem of purpose. A verified profile can still be a marketing vehicle. A verified identity can still be a brand. The badge confirms that the person is real. It does not confirm that the person is there for the reasons the platform claims to serve.

Where This Leaves Us

So here is the situation. The open market aggregates predators and strips away the social filters that used to catch them. The legacy platforms hide the same actors behind deniability and fake accounts by the hundreds of millions. The infrastructure for integrating verified profile onto social platforms is being built by governments around the world, but it solves authenticity without solving purpose. A real person can still be a predator. A verified account can still be a brand.

The fix has two halves, and they have to be built in order.

The first half is identity. Tie every account to a real, verified, state-backed identity — without handing the platform the data. The European Union is already building this: eIDAS 2.0, the EUDI Wallet, a portable credential that proves what needs to be proven and discloses nothing else and its local equivalents. The Anglo platforms are improvising their own walled-garden versions because they have to. The EU, Kenya, India and others have the actual infrastructure. The platforms have some patches.

The second half is community. Take the connection out of the global feed and into moderated digital bubbles — clubs, associations, purpose-driven spaces where membership is verified, norms are enforced, and reputation is at stake. The club is the filter the market lacks. The moderator is the accountability mechanism the algorithm cannot provide.

Identity without community gives you verified strangers. Community without identity gives you unverified neighbors. You need both. The first half is arriving on schedule. The second half has not been built.

The First Half in Europe: eIDAS and the Digital Identity Wallet

The European Union has been building the infrastructure to fix the authenticity problem for over a decade, and it is now arriving. The framework is eIDAS 2.0 — Regulation (EU) 2024/1183, which entered into force on 20 May 2024. It is not a dating app regulation. It is a digital identity regulation, and it is the most consequential piece of identity infrastructure the EU has ever built.

The core of the framework is the European Digital Identity Wallet (EUDI Wallet). It is not a single app. It is a technical and legal standard — a wallet that every Member State must offer to its citizens and residents by the end of 2026. Germany’s national wallet, d-you, is scheduled to launch on 2 January 2027. Denmark’s AltID launched in June 2026. The rollout is staggered, but the deadline is binding.

The architecture is what matters. A wallet is a personal digital container that holds identity data and electronic attestations of attributes — a verified name, a verified age, a verified nationality, a verified address. A user presents a single fact to a platform without disclosing the rest. The eIDAS framework explicitly requires that the wallet be accessible and usable for everyone, including people at risk of poverty and people without a fixed address. It is designed for inclusion, not for exclusion.

For dating apps, the implication is direct. The EUDI Wallet can verify that a user is over 18 without disclosing their birthdate. It can verify that a user is who they claim to be without handing over a passport scan. The European Commission’s own age-verification app, launched in 2025 and built on the eIDAS framework, allows a platform to receive a signed, cryptographically verifiable yes or no on an age threshold — nothing more. The user’s identity and activity remain protected from disclosure throughout the process. The user can present the proof without any data flowing to the verification provider, and without the platform learning the user’s identity. The EU has explicitly designed the system to prevent tracking on online platforms.

The legal hook is the Digital Services Act. The DSA requires very large online platforms — which include the major dating apps — to conduct annual risk assessments and put in place mitigation measures to address systemic risks. The EUDI Wallet is explicitly framed as a tool that social networks and video-sharing platforms can leverage to comply with their DSA age-gating obligations. The Commission has stated that the DSA aims to make the internet safer while preserving fundamental rights, and that the principle of online anonymity remains unchanged. The wallet does not break anonymity. It verifies the attributes that matter without revealing the person behind them.

Petition 1428/2024, submitted to the European Parliament by a Romanian citizen, goes further. It calls for mandatory identity verification for all social media users in the EU through KYC — Know Your Customer — procedures, the same standard that banks use to verify account holders. The petitioner argues that disinformation, manipulation, foreign propaganda, and fake accounts destabilize public discourse and undermine trust in democratic institutions. The petition was declared admissible on 2 October 2025. The Commission’s reply, received on 30 January 2026, noted that the DSA does not currently require mandatory identity verification for social media accounts, but it acknowledged the petition and the ongoing debate. The Commission also pointed to the EUDI Wallet as a tool that could facilitate such measures and promote trust and security in online interactions by enabling secure identification and authentication.

The Anglo Comparative: What the Platforms Are Doing Without the State

The Anglo platforms have not waited for regulation. They have built their own verification systems, and the European rollout is now a live test case.

Tinder expanded its ID verification option to France, Germany, Italy, the Netherlands, and Spain in April 2025. The process requires a video selfie and a valid driver’s license, passport, or national ID card. Tinder checks the validity of the ID document and verifies the birthdate to confirm the user is over 18. If the user completes both ID and photo verification, Tinder checks whether the face in the video selfie matches both the ID photo and at least one profile picture. Tinder’s Head of Trust and Safety described the expansion as an effort to make the apps the safest and most trusted way to meet new people.

Bumble rolled out ID verification in March 2025, using technology from Veriff. Users upload a government-issued ID and take a selfie for biometric comparison. The feature is voluntary, but verified users receive a badge, and other users can filter for ID-verified profiles and request that their matches complete the verification process. The feature is available in the US, UK, Australia, Canada, France, India, Ireland, Spain, Germany, Mexico, and New Zealand.

Hinge, also owned by Match Group, extended FaceTec liveness detection and facial age estimation to users in the UK and Australia in February 2026, to comply with the UK’s Online Safety Act and Australia’s Social Media Minimum Age Act. The UK legislation, which went into effect in July 2025, requires age verification for any site or app hosting content that might be restricted to adults.

The Anglo platforms are building verification because regulation is forcing them to, because users are demanding it, and because the liability of not building it is rising. But the verification they are building is document-based: upload an ID, take a selfie, wait for a badge. It is a snapshot. It is not a continuous, privacy-preserving, state-backed identity layer. It is a patch on a system that was never designed to verify anyone.

The difference between the eIDAS approach and the Anglo platform approach is the difference between infrastructure and improvisation. The EUDI Wallet is a state-backed, interoperable, privacy-preserving identity layer that every Member State must provide. It is a public good, designed to be used across every digital service, not just dating apps. The Anglo platforms are building walled gardens. Tinder’s verification works on Tinder. Bumble’s verification works on Bumble. The badge is platform-specific, the data is platform-controlled, and the privacy depends on the platform’s policies rather than on a legal framework that constrains what the platform can do with the data. The EU system is built on the principle that the platform should never have the data in the first place.

The Second Half: The Moderated Digital Bubble

The EUDI Wallet is the first half. It is the infrastructure that ties an account to a real identity without disclosing the identity. It is the state-backed layer that the Anglo platforms are trying to replicate with their own, inferior, walled-garden systems. It is the most ambitious identity project in the democratic world, and it is arriving on schedule.

But the wallet solves only the problem of authenticity. It verifies that the user is real. It verifies that the user is over 18. It does not verify that the user is there for the reasons the platform claims to serve. A verified predator is still a predator. A verified fraudster is still a fraudster. A verified user who has passed the platform’s baseline check is now free to behave as he wishes, with the added credibility of a blue badge.

The second half is the solution that no major platform is building, because it does not scale. Take the social connection out of the global feed and into moderated digital bubbles, moderated by clubs.

The logic is not new. It is the logic of the local digital space, which the Stanford Social Innovation Review describes as essential civic infrastructure. Most platforms prioritize metrics such as likes, comments, and time spent, incentivizing content that provokes strong reactions to keep users engaged, often at the cost of constructive dialogue. To counter this risk, local digital spaces rely heavily on community stewards — the volunteer moderators and organizers who embody the values of their local communities and work to create digital spaces that are welcoming and informative.

The club is the moderator. The club is the filter. A photography club, a climbing club, a book club, a neighborhood association, a professional guild. These are bounded, interest-based social worlds where status is earned through contribution, knowledge, reliability, and shared passion rather than through comparative attractiveness against an infinite pool. The mating market within these spaces operates on the same charm-looks-wits logic as all human pairbonding, but it is shielded from the algorithmic compression that renders most participants invisible and from the predatory concentration that the open architecture guarantees.

The moderation is the key. The club has a membership, and the membership can be revoked. The club has a reputation, and the reputation is at stake. The club has a purpose, and the purpose is not engagement metrics. The moderator is not an algorithm optimizing for time-on-site. The moderator is a person who knows the members, who understands the norms, and who has the authority to remove someone who violates them. This is not a perfect system. It is a system with a floor.

The Anatomy of the Moderated Bubble

What does a moderated digital bubble look like? It is not a Facebook Group with a volunteer admin who checks in once a week. It is a platform with a purpose, a membership, and a governance structure. The members are verified — by the EUDI Wallet, in the European case, which means the verification is portable, privacy-preserving, and not controlled by the platform. The purpose is defined. The norms are explicit. The moderation is active. The exit is possible. The bubble is small enough that reputation matters and large enough that the social graph does not saturate.

The bubble is also not a dating app. The dating app is a market for partners. The bubble is a community for people. The community is the point, and the partnership is a possibility that emerges from the community rather than a product that the platform sells. This is the inversion that matters. The dating app sells the search. The bubble sells the belonging. The search is where the predators concentrate. The belonging is where the predators are screened out by the community that knows them.

The evidence for the protective effect of community is indirect but consistent. The research on local digital spaces finds that well-designed spaces with well-supported stewards can keep communities connected and safe, while poorly moderated spaces devolve into arenas for shouting matches and harm. The research on dating app-facilitated violence finds that the absence of accountability is the mechanism: the platforms’ announcements are nothing more than PR because opt-in models do not prevent sexual violence and do not hold people accountable. The bubble is the accountability mechanism that the open market lacks.

The Technical Architecture: AT Protocol and the Fediverse

The infrastructure for the second half is already being built. It is not being built by the dating apps. It is being built by the open protocol movement.

AT Protocol — the protocol behind Bluesky — is a prime candidate for the universal public profile. It is an open, decentralized standard for public conversation that allows users to own their identity, their data, and their social graph across multiple applications and services. A user on AT Protocol has a persistent, portable identity that is not tied to any single platform. The profile is public. The graph is user-controlled. The data is portable. This is the identity layer that the second half needs: a verified, portable, user-owned profile that a club can accept without becoming a surveillance apparatus, and that a user can present without surrendering their data to a dating app.

The Fediverse — the network of interconnected servers running open protocols like ActivityPub — is a prime candidate for the clubs. The Fediverse is not a single platform. It is a federation of communities, each with its own norms, its own governance, and its own moderation. Mastodon, PeerTube, Lemmy, and the rest are not products. They are protocols. A club on the Fediverse is a server with a purpose, a membership, and a governance structure. The moderator is accountable to the members, not to an engagement algorithm. The community is the point. The relationship is the byproduct.

The combination is what the second half requires. AT Protocol provides the portable, user-owned identity. The Fediverse provides the moderated, purpose-driven community. The profile is verified. The community is governed. The predator has no place to hide, because the identity is persistent and the community is accountable. The club knows who you are, because your identity travels with you across every server you join. The club can remove you, because the membership is revocable. The club can report you, because the identity is public and the reputation is at stake.

This is not a theoretical architecture. It is being built. AT Protocol is live. The Fediverse is live. The protocols exist. What does not exist is the integration — the portable profile that a Fediverse server can accept as a membership credential, and the club structure that a user can join without abandoning the identity they have built elsewhere. That integration is the next step.

The Half-Measure and the Whole

The dating app has done what it can do. It has tied the account to a real identity. It has built the verification infrastructure. It has implemented the safety features. The half-measure is not a failure. It is a prerequisite. But it is only half.

The other half is the bubble. The moderated community. The club. The space where the identity is verified, the purpose is shared, and the norms are enforced. The dating app is a market. The bubble is a village. You cannot find a partner in a market without being exposed to the market’s predators. You can find a partner in a village, because the village knows who is who.

The EU has built the infrastructure for the first half. It has not built the second. The EUDI Wallet is the identity layer that makes the bubble possible — a portable, privacy-preserving verification that a club can accept without becoming a surveillance apparatus, and that a user can present without surrendering their data to a platform. AT Protocol is the portable profile that carries the verification across every server. The Fediverse is the community infrastructure that makes the clubs possible. The technology exists. The legal framework exists. The protocols exist. The only thing missing is the will to build the integration that makes the whole architecture work.

---

References

AAP News. (2024, November 15). Dark side of dating with some app users more vulnerable. AAP News. https://aapnews.aap.com.au/news/dark-side-of-dating-with-some-app-users-more-vulnerable

Biometric Update. (2025, March 19). Bumble adds Veriff ID verification. Biometric Update. https://www.biometricupdate.com/202503/bumble-adds-veriff-id-verification

Bundestag. (2026, October 14). Digitalausschuss stimmt für Gesetz für digitale Identitäten. Deutscher Bundestag. https://www.bundestag.de/presse/hib/kurzmeldungen-1224792

ComplyCube. (2024, June 21). The catfishing crisis and social media identity verification. ComplyCube. https://www.complycube.com/en/the-catfishing-crisis-and-social-media-identity-verification/

Council of the European Union. (2026, June 9). European business wallets: Council adopts negotiating position. Consilium. https://www.consilium.europa.eu/en/press/press-releases/2026/06/09/european-business-wallets-council-adopts-negotiating-position/

eIDAS Pro. (2026, July 7). How to comply with Google Play’s new dating app age rules using eID. eIDAS Pro. https://eidas-pro.com

European Parliament. (2025). Petition No 0733/2025 by M.O. (Polish) on a mandatory human identity verification for all social media account accessible in the EU. PETI-CM-784281_EN.pdf. https://www.europarl.europa.eu/doceo/document/PETI-CM-784281_EN.pdf

European Parliament. (2026, January 30). Petition No 1428/2024 on the introduction of mandatory identity verification for all social media users in the EU. PETI-CM-784281_EN.pdf. https://www.europarl.europa.eu/doceo/document/PETI-CM-784281_EN.pdf

EUR-Lex. (2025, October 10). Commission communication on age verification. Official Journal of the European Union, C 2025/05519. https://eur-lex.europa.eu/legal-content/HU/TXT/?uri=OJ:C_202505519

GIJN. (2025, June 5). 调查约会应用在保护女性上系统性失效，他们是如何做到的？ GIJN. https://zh.gijn.org/stories/investigating-systemic-failure-enabling-abuse-dating-apps/

Lexology. (2026, March 27). European Union: EUDI Wallet harmonizes identification and age gating. Lexology. https://www.lexology.com/library/detail.aspx?g=4f41b246-e7f3-4570-a3ed-f509b5536b28

Malwarebytes. (2026, September 25). LinkedIn adds new checks for fake profiles and work histories. Malwarebytes. https://www.malwarebytes.com/blog/news/2026/09/linkedin-adds-new-checks-for-fake-profiles-and-work-histories

NDTV. (2026, September 17). Dating app scams use gig workers, AI chatbots to deceive thousands. NDTV. https://www.ndtv.com/artificial-intelligence/dating-app-scams-use-ai-chatbots-and-real-gig-workers-to-deceive-thousands-of-users-12058729/amp/

Signicat. (2026, June 1). Identity verification for online dating: Why eIDs and NFC are the new standard for trust. Signicat. https://www.signicat.com/blog/identity-verification-for-online-dating-why-eids-and-nfc-are-the-new-standard-for-trust

Stanford Social Innovation Review. (2025, March 4). Local digital spaces are essential civic infrastructure. SSIR. https://ssir.org/articles/entry/investing-in-digital-civic-infrastructure

Stickley on Security. (2026, January 24). Man vs. woman: Romance scams and finding love online. Stickley on Security. https://www.stickleyonsecurity.com/man-vs-woman-romance-scams-and-finding-love-online/

TechCrunch. (2025, March 17). Bumble heightens safety measures with new ID verification feature. TechCrunch. https://techcrunch.com/2025/03/17/bumble-heightens-safety-measures-with-new-id-verification-feature/

TechCrunch. (2025, October 22). Tinder will require new users in the US to verify their identity with a selfie. TechCrunch. https://techcrunch.com/2025/10/22/tinder-will-require-new-users-in-the-us-to-verify-their-identity-with-a-selfie/

The Associated Press. (2024, October 1). Australia’s online dating industry adopts code of conduct to keep users safer. AP News. https://hosted.ap.org/listenupyall/article/a6cba10c81e60f1cb8866a2a1201c163/australias-online-dating-industry-adopts-code-conduct-keep

Tinder Pressroom. (2025, April 4). Tinder® kündigt Erweiterung der ID-Verifizierung für Nutzer:innen in Frankreich, Deutschland, Italien, den Niederlanden und Spanien an. Tinder Pressroom. https://de.tinderpressroom.com/idv2025

Yahoo Tech. (2026, September 14). Fake dating apps used Claude to scam 25,000 people, Anthropic says. Yahoo Tech. https://tech.yahoo.com/ai/claude/articles/fake-dating-apps-used-claude-163457117.html

Youverse. (2026, August 10). eIDAS 2.0 and age assurance: Age as a reusable attribute. Youverse. https://www.youverse.id