Match Group's AI Dilemma: Competing with Companions It Can't Dismiss
Key Points
- •Match Group saw its share price drop from $182 in 2021 to around $35, whilst paying users fell 6% to 10.3 million.
- •Replika reported a 35% year-on-year subscriber growth in 2023, charging $69.99 annually for romantic conversation modes.
- •Character.AI claimed over 20 million users while charging a $9.99 monthly subscription for premium access features.
- •Match Group Head of Asia told an industry conference that relationships with artificial intelligence will always feel empty compared to human connection.
Match Group's Head of Asia has fired a warning shot across the bow of AI companionship apps, telling an industry conference that relationships with artificial intelligence will 'always feel empty' compared to human connection. Bernard Kim positioned Match's extensive AI investments as tools to enhance rather than replace human dating—a distinction the company is increasingly anxious to draw. The comments arrive as AI companion platforms report user numbers that would make most dating apps envious, raising existential questions for an industry watching its market share erode.
This isn't a philosophical debate about the nature of connection—it's Match Group watching a competitor category emerge that doesn't play by dating's rules. AI companions don't have chicken-and-egg marketplace problems, don't deal with gender ratio imbalances, and don't lose users to successful matches. They're SaaS businesses disguised as relationship apps, and they're growing whilst Match's monthly active users stagnate.
Kim's comments read less like confident dismissal and more like the sound of an industry leader realising the threat might not come from Bumble or Hinge.
The challenge for Match Group is that its critique rings hollow against its own product roadmap. The company has spent the past 18 months embedding AI throughout its portfolio—Tinder's 'Photo Selector' uses computer vision to optimise profile pictures, Hinge deploys machine learning for 'Most Compatible' matches, and Match.com now offers AI-powered conversation starters. CEO Bernard Kim (no relation to the Asia executive) told investors in February that AI would be 'transformational' for the business, pledging continued investment in recommendation algorithms and automated matching.
Create a free account
Unlock unlimited access and get the weekly briefing delivered to your inbox.
Where Match draws the line, according to its Asia leadership, is at AI replacing the human on the other end of the conversation. The company frames its technology as connective tissue between real people—optimising introductions, smoothing friction points, perhaps drafting that first message. What it won't countenance, at least publicly, is the model pioneered by Replika and Character.AI: the AI as the relationship itself.
The economics of never being ghosted
The business model contrast is stark. Traditional dating apps monetise through a combination of subscriptions for enhanced features and à la carte purchases for boosts and super likes. Match Group reported average revenue per payer of $15.93 in Q4 2023, down from $16.24 the previous year. Conversion rates from free to paid hover around 3-4% for most apps. The model depends on dissatisfaction—users pay because they're not getting enough matches, messages, or dates.
AI companions flip this entirely. Replika charges $69.99 annually for 'Replika Pro', which unlocks romantic and intimate conversation modes. Character.AI offers a $9.99 monthly subscription for faster response times and early access to features. Crucially, there's no matching algorithm to game, no competition for attention, no risk of rejection. The product delivers guaranteed engagement, 24/7 availability, and conversations calibrated to user preferences through large language models.
Retention economics tell an uncomfortable story for traditional dating. According to data from Sensor Tower, Replika's subscriber base grew 35% year-on-year in 2023, whilst Match Group's paying users declined 6% to 10.3 million. Anecdotal evidence from Reddit and user communities suggests some AI companion users are former dating app subscribers who grew tired of the traditional app experience—the endless swiping, the conversations that fizzle, the dates that disappoint.
Match Group's defensive positioning relies on the assumption that AI relationships represent a fundamentally inferior good—a stopgap for the lonely rather than a genuine alternative to human partnership.
Kim's assertion that AI companionship feels 'empty' might resonate with those who've never tried it, but user testimonials from AI companion platforms suggest a more complicated reality. Users report feeling heard, understood, and emotionally supported by their AI partners. Whether these feelings are 'real' is a philosophical question; whether they're commercially viable is answered by the subscription data.
The uncomfortable middle ground
What makes Match Group's position particularly awkward is that it's already halfway down the slippery slope it's trying to avoid. The company's AI features aren't just optimising matches—they're increasingly generating the interactions themselves. Tinder's conversation prompts use natural language processing to suggest context-specific opening lines. Several Match properties now offer AI-assisted profile writing. The boundary between AI-as-tool and AI-as-intermediary blurs with every product update.
The industry's larger worry extends beyond direct competition from companion apps. If users become comfortable with AI-mediated emotional connection, the value proposition of traditional dating—meeting real humans for potential relationships—starts to look expensive, time-consuming, and high-risk by comparison. Dating apps have always competed against alternatives like meeting through friends, at work, or staying single. They now face a competitor that offers connection without vulnerability, intimacy without logistics, and companionship without compromise.
Regulatory winds might offer Match some defence. The EU is examining AI companion apps under the Digital Services Act (DSA) framework, particularly regarding their impact on minors and vulnerable users. The UK's Online Safety Act (OSA) could theoretically extend to apps that facilitate 'relationships' with AI entities, though the legislation primarily targets user-generated content rather than synthetic interactions. Match Group, with its extensive compliance infrastructure and experience navigating relationship-focused regulation, would likely welcome a framework that imposes costly requirements on AI companion startups.
The coming 12 months will test whether Match Group's stance represents strategic clarity or denial. The company reports Q1 earnings in early May—watch for any commentary on competitive dynamics beyond traditional rivals. If AI companion apps continue their user growth trajectory whilst dating app engagement stagnates, Match Group's philosophical objections to the category may prove insufficient protection against the market's judgment. Meanwhile, other dating executives have described AI chatbot relationships as 'playing with fire', suggesting the industry recognises the threat even as it deploys AI as a 'wingman' across its own platforms.
Key Takeaways
- •Artificial intelligence companion apps operate as Software-as-a-Service models that avoid the traditional marketplace imbalances and user churn inherent in online dating platforms.
- •Emerging regulatory frameworks like the European Union Digital Services Act could benefit Match Group by imposing expensive compliance requirements on companion startups.
- •Dating app operators risk undermining their human-centric value propositions as internal artificial intelligence features shift from matching assistance to automated conversation generation.
Frequently Asked Questions
Financial Intelligence Desk
The DII Financial Intelligence Desk covers earnings, valuations, funding and the financial performance of the global online dating industry.
Comments
Join the discussion
Industry professionals share insights, challenge assumptions, and connect with peers. Sign in to add your voice.
Your comment is reviewed before publishing. No spam, no self-promotion.
