Chatfishing: When AI Writes Your Messages on Dating Apps

What Is Chatfishing?

Chatfishing is the practice of using language models like ChatGPT or Claude to draft messages on dating apps. Users copy and paste real conversations into these tools, asking for replies, error analysis, or opening lines. According to The Next Web, Google searches for related terms have increased 5,000%, and 26% of American adults (and 49% of those under 30) admit to having done it. This phenomenon is not entirely new: traditional 'catfishing' (using fake identities) has existed since the early internet, but AI takes it to another level. Now you don't need a stolen photo; just a prompt for a model to generate perfect responses, eliminating human effort and authenticity.

Why Does It Matter?

This phenomenon raises deep questions about authenticity, consent, and the role of AI in human relationships. If someone believes they are connecting with another person but is actually interacting with a bot trained on someone else's data, the basic trust of any bond is broken. Moreover, dating apps like Tinder and Hinge have stated they cannot detect or block this practice, opening the door to massive, unregulated use. On social media like Reddit and TikTok, users compete to show the best 'prompts' and results, creating a culture of seduction optimization reminiscent of 2000s 'pick-up artists,' but now with algorithmic backing. The impact goes beyond the individual: if AI becomes the standard intermediary for courtship, spontaneity and vulnerability—pillars of human connection—could disappear.

Consequences for the Market and Society

• For users: the risk of disappointment and emotional fraud increases. Conversations can be perfect, but the real meeting reveals an insurmountable gap. Additionally, there is the danger that data shared with AI (likes, fears, intimate experiences) may be stored and potentially used without explicit consent, as has already happened with other generative AI services.
• For platforms: they lose their unique value proposition (genuine human connection) and face a crisis of trust. They could implement authenticity checks—like real-time CAPTCHA tests or writing pattern analysis—or limit the use of AI APIs. However, these measures would be difficult to apply without violating user privacy. A historical precedent is social networks facing bots: Facebook and Twitter took years to develop effective detection systems, and still struggle with automated accounts today.
• For the AI industry: a controversial use case emerges that could accelerate debates on ethics and regulation. Companies like OpenAI and Anthropic already face criticism for allowing deceptive uses of their models. If chatfishing scales, lawsuits for emotional damages or fraud could arise, similar to what happened with romance scams on platforms like Match.com. Moreover, the phenomenon could drive the creation of 'AI detectors for AI' in conversations, a niche already explored by startups like Sensity (specializing in deepfakes).

What Should Readers Know?

Chatfishing is not illegal, but it is deceptive. If you use AI to flirt, be honest about it. If you suspect your date is doing it, watch for inconsistencies (overly polished language, generic responses, lack of human errors like typos or filler words). Apps likely won't protect you, so responsibility falls on each individual. On a societal level, this phenomenon forces us to rethink what 'authentic connection' means in the age of AI. Just as the arrival of smartphones transformed dating with apps like Tinder, generative AI now promises a new revolution—but perhaps not in the direction we expected. The key question is: to what extent will we delegate our intimacy to algorithms? Chatfishing is just the tip of the iceberg.

«26% of adults in the US already use AI to write on dating apps. Authenticity is at stake.»