Dating Apps Face Unavoidable Friction: Regulatory Pressure Mounts

Industry leaders gathering at Tech Show London last week delivered an uncomfortable message to dating operators: the apps are going to have to get harder to use. Speaking on a panel about romance fraud, executives from ID verification firm Onfido, fraud detection company Sardine, and the National Crime Agency argued that mandatory ID checks, AI-powered deepfake detection, and other 'friction' measures are now essential to combat scammers who've become too sophisticated for human users to identify. The consequence is apps that feel slower, clunkier, and less frictionless than the product experiences that drove user growth for the past decade.

The framing is blunt but correct. According to the panel, romance scammers are now deploying deepfake video and voice technology that has advanced from laughably detectable to near-convincing in roughly two years. Simon Newman, managing director at Onfido, told attendees that 'human capabilities are diminishing' in spotting fake profiles and manipulated media—not because users are becoming less vigilant, but because the technology is advancing faster than human detection skills can adapt.

That shifts the burden squarely onto platforms to implement detection systems that most users will experience as friction: additional verification steps, delays in profile activation, and automated checks that may flag legitimate accounts.

Create a free account

Unlock unlimited access and get the weekly briefing delivered to your inbox.

Register free

No spam. No password. We'll send a one-time link to confirm your email.

The regulatory clock is ticking

Legislative pressure is no longer a future concern. The UK's Online Safety Act, which came into force incrementally throughout 2024, already places responsibility on platforms to prevent fraud, with Ofcom empowered to impose fines up to 10% of global revenue for non-compliance. Damien Boyle, anti-fraud lead at the National Crime Agency, explicitly referenced the Act during the panel, framing it as leverage to push platforms toward mandatory verification.

That timeline matters: trust and safety teams at UK-facing operators are already working to interpretation of 'proportionate measures', and ID verification sits squarely within the anticipated compliance framework. Across the Atlantic, momentum is building for federal legislation. Congress is considering a Romance Scam Prevention Act that would require platforms to notify users when their accounts are flagged by fraud detection systems.

The bill hasn't passed yet, but the direction is clear: regulators in major markets are moving toward mandating the very friction measures the industry is debating. For Match Group (MTCH), Bumble (BMBL), and other multi-jurisdiction operators, the compliance matrix is becoming more complex and more expensive.

The cost implications are significant. Soups Ranjan, co-founder of fraud detection firm Sardine, noted on the panel that implementing AI-driven deepfake detection and continuous monitoring systems requires substantial investment in both technology infrastructure and the operational capacity to handle false positives. Bumble disclosed in its Q3 2024 earnings that trust and safety headcount had increased 23% year-over-year, contributing to elevated operating expenses.

The deepfake arms race

What's changed in the past 24 months isn't the existence of romance scams, but the tools available to execute them. Deepfake video and voice synthesis have moved from niche research projects to consumer-accessible apps capable of generating convincing real-time video calls. Newman described this as a shift from human-versus-human vigilance to AI-versus-AI detection battles.

That fundamentally alters the economics of scamming: where previously a fraudster needed to invest time building trust through text conversations, they can now deploy automated systems that mimic video interactions at scale. The panel suggested that introducing friction—mandatory selfie verification, delayed profile activation, flagging accounts that attempt to move conversations off-platform—could make scams 'much more time-consuming' for fraudsters, reducing the economic incentive.

The same features that made dating apps successful are now the vulnerabilities regulators are targeting.

That's the theory. In practice, the effectiveness of friction depends on implementation. Heavy-handed verification that flags too many false positives risks alienating legitimate users, particularly those in demographics already wary of sharing government ID with dating platforms. Light-touch friction that scammers can bypass cheaply won't materially change the economics.

There's an uncomfortable irony here. The pandemic surge in online dating—which saw Tinder hit 75 million monthly active users by mid-2021 and normalised rapid digital intimacy—created precisely the conditions scammers exploit. Platforms optimised for speed and ease of connection, with minimal barriers to profile creation and messaging, because that's what drove growth. The UK Serious Fraud Office reported romance fraud losses of £92.3M in 2023, up 27% from 2021.

What operators should be watching

The shift toward mandatory friction is happening unevenly across markets and platforms. Bumble introduced selfie verification in 2020 and claims a 35% adoption rate as of Q4 2024, but it remains optional. Tinder began testing video verification in select markets in 2023 but hasn't rolled it out globally. Grindr (GRND) has resisted ID verification, citing privacy concerns particularly relevant to its LGBTQ+ user base.

That fragmentation won't survive sustained regulatory pressure. Operators delaying investment in verification infrastructure are increasing compliance risk without materially reducing costs—they'll still need to build or buy the technology, just under a tighter timeline and with potential enforcement action looming.

The deeper strategic question is whether friction can be implemented in ways that feel like protection rather than inconvenience. Ranjan suggested during the panel that platforms could reframe verification as a trust signal, making verified profiles more visible or adding badges that signal legitimacy. That's product design attempting to turn a liability into a feature.

Whether users perceive additional steps as valuable protection or annoying gatekeeping will determine how much friction the market can absorb before growth stalls. For an industry already grappling with user fatigue and slowing subscriber growth—MTCH reported paid subscriber declines in three of the past four quarters—adding friction to the onboarding experience is a material risk to acquisition funnels.

The panel's message is politically inconvenient but operationally unavoidable: dating apps are going to get harder to use because the alternative—continuing to rely on human vigilance against AI-generated deception—has already failed. Operators can either design that friction intentionally, with an eye toward minimising user impact, or wait for regulators to mandate it on timelines and terms they don't control.