Dating Apps' AI Obsession: Users Aren't Buying It

Dating platforms have built their businesses on understanding what their users want. A new survey suggests they may have got it spectacularly wrong on artificial intelligence. According to research from Instabug, dating app users are the least receptive to AI features of any app category surveyed, creating an uncomfortable gap between industry investment and user appetite.

Just 15% expressed enthusiasm about AI integration, whilst 63.5% said AI capabilities wouldn't influence their choice of platform at all. The remaining 36.5% indicated they would consider switching apps based on AI features: a figure that's either reassuringly high or worryingly low, depending on whether you're building the feature or funding it.

The findings arrive at an inconvenient moment. Match Group (MTCH) has already committed to AI across its portfolio, with CEO Bernard Kim describing artificial intelligence as central to the company's product roadmap in the Q3 2024 earnings call. Bumble (BMBL) has been similarly bullish, positioning its AI-powered 'Opening Moves' feature and photo verification tools as key differentiators. Grindr (GRND) has rolled out AI-powered chat suggestions.

The commitments are public, the engineering resources allocated, the features already shipping.

The DII Take

The industry is walking into a trap of its own making. Dating operators have convinced themselves that AI is a product imperative when the data suggests it's more likely a distraction from the core problems users actually want solved: better matches, fewer time-wasters, and protection from harassment.

The 63.5% who say AI won't influence their decision aren't confused: they're telling you it's not the feature they're asking for. Platforms that double down on AI at the expense of trust and safety improvements aren't innovating; they're just spending money in the wrong place.

The gender gap that matters

The survey revealed a 10-percentage-point split between male and female users. Twenty per cent of men see AI as beneficial to their dating app experience, compared with just 10% of women. That gap should concern every operator reading this.

Women are typically the more selective user base, the demographic dating platforms work hardest to retain, and the group that faces the most acute safety concerns. They're also the users most likely to disengage when platform quality deteriorates. If AI features are twice as appealing to men as to women, platforms risk accelerating the gender imbalance that already plagues the industry, particularly on heterosexual platforms where user ratios directly impact match quality and retention.

The scepticism isn't surprising. Women on dating platforms already contend with a higher volume of unwanted contact, more aggressive messaging, and greater privacy concerns. AI-generated openers, match predictions, and conversation suggestions don't solve those problems. They make the experience feel more synthetic, less authentic, and potentially less safe if users can't distinguish between genuine engagement and algorithmic prompting.

Platforms have historically justified feature development by pointing to engagement metrics and power user behaviour. But AI features introduce a different calculation. If 10% of women actively want AI and 63.5% of all users say it won't influence their choice, the addressable market for these features is vanishingly small, and concentrated among a demographic that platforms are already over-serving.

When innovation becomes liability

The 36.5% of users who said they'd consider switching apps for AI features represents the industry's best-case scenario. That cohort, early adopters, tech-forward users, those willing to experiment, is the audience platforms are betting on. The gamble is that this minority will validate the technology, shift broader sentiment, and create competitive pressure that forces holdouts to adopt.

It's not an unreasonable bet. Dating apps have successfully introduced features that initially seemed invasive or unnecessary: photo verification, video profiles, voice notes. Users adapted. Adoption followed. The platforms that moved first gained an edge.

But AI is different. Those earlier innovations solved visible problems: catfishing, low-effort profiles, scammers. They increased trust. AI features, as currently conceived, are solutions in search of problems.

Auto-generated conversation starters don't make matches better; they make them easier, which isn't the same thing. Algorithmic match predictions don't address the core complaint that apps aren't showing users compatible people: they just add a layer of technological mystique to a fundamentally unsolved problem.

The risk for platforms is that AI becomes associated with inauthenticity. Dating app fatigue is already a documented phenomenon, driven by the sense that apps prioritise engagement over outcomes.

If users come to believe that the person they're chatting with is using AI to craft responses, or that matches are being surfaced based on opaque algorithmic predictions rather than genuine compatibility, trust erodes further. Dating app fatigue is already a documented phenomenon, driven by the sense that apps prioritise engagement over outcomes.

Match Group and Bumble can absorb the development costs. Both have the engineering talent, the data infrastructure, and the balance sheets to experiment. What they can't afford is a perception shift that positions AI as the enemy of genuine connection, particularly among female users, who have options and aren't shy about exercising them.

The methodology gap

The Instabug survey doesn't disclose sample size, demographic breakdown beyond gender, or geographic distribution. That's a problem. User sentiment on AI varies significantly by age, market maturity, and cultural context. A survey skewed towards younger, tech-literate users in developed markets would yield very different results than one reflecting the global dating app user base.

Without that context, the 15% enthusiasm figure is directionally useful but not definitive. Platforms will rightly argue that their own user research, proprietary, longitudinal, and segmented, provides a clearer picture. But the broad pattern is harder to dismiss: dating app users are categorically less interested in AI than users of other app categories. That's not a methodology problem. That's a signal.

The challenge for operators is deciding whether to follow the data or lead the market. Platforms that pull back on AI risk being outflanked by competitors willing to take the bet. Those that push forward risk alienating the majority who've already said they're not interested. The worst outcome is the middle path: shipping AI features that satisfy neither the sceptics nor the enthusiasts, burning capital and engineering time on functionality that ends up ignored.

The industry has bet heavily that AI is the next frontier. The users, it seems, are still waiting to be convinced.