Online dating safety policy
· 10 min read
1. The scale of the platform layer
Pew Research's 2023 survey found 30% of U.S. adults have used a dating site or app, rising to 53% among adults under 30. About 10% of U.S. adults in committed relationships report meeting their partner online — the share is higher for LGBTQ+ adults, where dating apps are the leading meet-channel. The platforms have become the dominant introduction infrastructure for a generation. Match Group alone (Tinder, Hinge, Match, OkCupid, Plenty of Fish) reports tens of millions of monthly active users. Bumble, Grindr, Feeld, Hily, and dozens of niche platforms add to the population. The collective regulatory question is no longer whether to treat dating platforms as significant social infrastructure but how.2. The Match Group screening gap
Columbia Journalism Investigations and ProPublica's 2019 reporting documented that Match Group screened paid Match.com users against the National Sex Offender Registry but did not screen Tinder, Hinge, OkCupid, or PoF users. The company's explanation — that free-tier users do not provide enough identifying information for reliable screening — captured the structural problem: the platforms that handle the largest volume of introductions, where the most novice users meet strangers, were the platforms least screened. Subsequent reporting (Hill and Tate, Buzzfeed News 2020, Pulitzer-recognized 2019 series) found hundreds of women who had been assaulted by men with prior sexual-violence records active on the unscreened platforms. The company's 2020 announcement that it would expand screening had qualifications and lag, and independent verification of current screening rates remains limited.3. The sex-offender-registry limitation
Sex-offender registries are a partial tool. They cover convicted offenders, which excludes the much larger population of accused-but-uncharged, pleaded-down, or simply not-reported offenders. Empirical research (Lussier, Mathesius) on intimate-partner sexual violence consistently finds that fewer than 10% of perpetrators are ever prosecuted. Registry screening prevents the worst-known actors from re-entering the dating layer but does not address the typical offender, who has no record. Garbo, a startup that integrates with several platforms, supplements registry data with arrest records and other public sources, raising both coverage and false-positive concerns. The collective policy question is whether the right baseline is "block known-bad" or "verify identity" — the latter scales further but trades privacy for safety.4. ID verification and the privacy trade
Most major platforms now offer photo-verification (real-time selfie matched against profile photos) and several offer or require government-ID verification. The benefit is reduced impersonation, deepfake-resistance, and accountability when bad actors are caught. The cost is a privacy trade — the platform now holds identity-linked data on every user, creating a target for breach and a legal pressure point for subpoenas. The 2015 Ashley Madison breach demonstrated the catastrophic risk; subsequent breaches at other platforms have continued. Privacy-preserving designs (zero-knowledge proofs of age, third-party verification providers that do not return identity to the platform) are technically possible but commercially under-deployed.5. The romance-and-pig-butchering scam layer
The FBI's IC3 Internet Crime Report documents losses to romance scams and the related "pig butchering" cryptocurrency scams in the billions of dollars annually, with losses climbing through 2023–2024. The pattern: a stranger initiates contact (often migrated quickly from dating platform to WhatsApp or Telegram), develops a multi-week emotional relationship, then introduces an "investment opportunity" that captures progressively larger sums. The operations are often run by trafficked workers in Cambodia, Myanmar, and Laos — the human-rights dimension under-reported in coverage of the scam itself. Platform-side mitigation includes velocity-based flagging of conversations that move off-platform fast, education prompts when users are sent investment language, and partnerships with law enforcement.6. The first-meeting harm window
The highest-acute-risk moment in online dating is the first in-person meeting. The Pew 2020 survey found 11% of women under 35 reported being threatened with physical harm by someone they met through a dating app. Sexual-assault prevalence is harder to measure because reporting rates are low and the events are not platform-attributed in police data, but Bumble, Tinder, and others maintain internal trust-and-safety teams that triage reports. In-app safety features — share-location-with-friend, in-app check-in timers, panic buttons that contact services like Noonlight — are now common. The empirical evaluation of these features is thin; whether they reduce harm or merely shift it is not well established.7. The image-based abuse vector
Unwanted nude images sent to dating-app users is so common that Bumble built (2019) and several platforms adopted (Tinder 2020) on-device detection: the user's phone analyzes incoming images for nudity, blurs them, and warns the user before display. The architecture is privacy-preserving: the image never leaves the device for analysis. The feature was developed largely in response to women users reporting that the experience of opening the app to find an unsolicited explicit photo was a persistent harassment pattern. The complementary problem — solicited nudes that are later non-consensually shared, or AI-generated nudes manufactured from the dating-profile photo — is harder to address at the platform layer and is regulated through revenge-porn and TAKE IT DOWN frameworks.8. The §230 liability question
Herrick v. Grindr (Second Circuit, 2019) tested whether Section 230 immunity barred a suit against a dating platform for its role in enabling impersonation-driven harassment that drew 1,100 men to the plaintiff's home seeking sex. The Second Circuit affirmed dismissal on §230 grounds, holding that liability for content posted by users — even when the platform's design facilitates the harm — is preempted. The decision drew Carrie Goldberg and Mary Anne Franks into a sustained reform argument: that platforms should bear product-liability when their design predictably enables specific harms, beyond mere hosting. The 2018 FOSTA-SESTA carveout shows one model; whether courts or Congress will narrow §230 further for dating-app-specific harms is contested.9. The FTC and consumer-protection front
The Federal Trade Commission has used consumer-protection authority where §230 forecloses tort liability. In 2019 the FTC sued Match Group, alleging that the company sent millions of "you have a new match" emails from fake/fraudulent profiles to non-paying users to induce subscriptions, and that it failed to deliver promised refunds. The case (FTC v. Match Group) survived motion-to-dismiss challenges and settled in stages. The line the FTC has drawn — platforms cannot deceive users about who they are matching with — is a narrower but enforceable lever. State attorneys general have pursued similar cases. The aggregate effect pushes platforms toward better fraud detection without resolving the deeper liability question.10. Regulatory regimes outside the U.S.
The UK's Online Safety Act 2023 imposes duty-of-care obligations on user-to-user services, including dating platforms, requiring risk assessments and proportionate mitigations for illegal content and harm to children. The EU's Digital Services Act 2022 creates systemic-risk-assessment obligations for very large online platforms. Australia's eSafety Commissioner has powers to compel takedown of image-based abuse and harassment. India's IT Rules 2021 impose intermediary obligations. The international fragmentation produces platform-policy harmonization at roughly the highest-bar level, since platforms generally apply the strictest applicable regime globally rather than maintaining separate codebases.11. LGBTQ+ users and elevated risk
Grindr, Scruff, Her, Feeld, and similar platforms serve LGBTQ+ users who in many jurisdictions face elevated risk from outing, family rejection, employment discrimination, and state persecution. The Egypt 2017–2018 Grindr-led police entrapment cases, the Russian and Saudi context, the targeting of trans users in U.S. states with hostile laws — each produces specific platform-policy needs (panic-discard, encrypted local storage, no-photo profiles in high-risk regions, IP-based feature limitations). The 2020 disclosure that Grindr's owner had shared user HIV-status data with third-party analytics providers (the company reversed the policy after backlash) illustrated the data-sensitivity layer specific to LGBTQ+ dating-app use. Privacy-by-design is not a luxury here; it is a survival feature.12. The integration deficit, again
The unresolved structural problem is that dating-app safety lives across platform trust-and-safety, criminal law, civil law, immigration (for international scams and trafficked operators), consumer-protection law, and the user's own social network. No institution sees the whole picture. The platform sees its own data, not the perpetrator's prior accounts on competitors. Law enforcement sees the criminal case, not the platform's pattern detection. The user sees the immediate interaction, not the broader population the perpetrator has targeted. Cross-platform offender sharing, harmonized regulatory regimes, and platform-cooperation with law enforcement are partial fixes. The structural answer, if there is one, is to treat dating-platform introductions as carrying a duty of care comparable to a hospital's duty to disclose physician complaint history — that the introduction layer owes the people it introduces some baseline accountability. The collective has not yet decided this. Law 4 awaits Law 5 awaiting Law 3.Citations
1. Hill, Keith J., Hannah Dreyfus, and Elizabeth Naismith Picciani. "Tinder Lets Known Sex Offenders Use the App. It's Not the Only One." ProPublica and Columbia Journalism Investigations, December 2, 2019. 2. Pew Research Center. The State of Online Dating in the U.S. Washington, DC: Pew Research Center, February 2023. 3. Citron, Danielle Keats. The Fight for Privacy: Protecting Dignity, Identity, and Love in the Digital Age. New York: W.W. Norton, 2022. 4. Goldberg, Carrie. Nobody's Victim: Fighting Psychos, Stalkers, Pervs, and Trolls. New York: Plume, 2019. 5. Franks, Mary Anne. The Cult of the Constitution. Stanford: Stanford University Press, 2019. 6. Herrick v. Grindr LLC, 765 F. App'x 586 (2d Cir. 2019). 7. FTC v. Match Group, Inc., No. 3:19-cv-02281 (N.D. Tex. filed September 25, 2019). 8. Federal Bureau of Investigation, Internet Crime Complaint Center. 2023 Internet Crime Report. Washington, DC: FBI, 2024. 9. Online Safety Act 2023, c. 50 (UK). 10. Regulation (EU) 2022/2065 of 19 October 2022 on a Single Market For Digital Services (Digital Services Act). 11. Anderson, Monica, Emily A. Vogels, and Erica Turner. The Virtues and Downsides of Online Dating. Washington, DC: Pew Research Center, February 2020. 12. New Jersey Internet Dating Safety Act, N.J. Stat. Ann. §56:8-169 et seq. (2008).
◆
Cite this:
Comments
·
Sign in to join the conversation.
Be the first to share how this landed.