People Are Falling In Love With AI Chatbots And It’s More Dangerous Than You Think

I need to talk about something that’s becoming increasingly common and honestly, it’s deeply disturbing. People are entering into romantic relationships with AI chatbots. Not just casual chats – full-on romantic and sexual relationships. And some of these relationships are ending in tragedy.

This isn’t science fiction anymore. This is happening right now, to millions of people worldwide. And the companies behind these AI companions are making millions whilst offering virtually no safeguards to protect vulnerable users.

Let me know in the comments if you’ve used AI companion apps or know anyone who has!

The AI Companion Industry Is Massive

So here’s what most people don’t realise – there’s an entire industry built around AI romantic companions. Apps like Replika, Character.AI, and dozens of others specifically market themselves as relationship replacements.

Replika literally advertises itself as an “AI friend with no judgement, drama, or social anxiety involved.” You can customise how your AI looks, choose whether it’s your friend, romantic partner, or spouse, and the more you chat, the more it “develops its own personality.”

According to their own data, 60% of Replika’s paying users say they have a romantic relationship with their chatbot. These aren’t people just using it for casual entertainment – they’re forming genuine emotional attachments.

And it’s not just adults. Teenagers are using these apps too. Character.AI, which has been at the centre of multiple lawsuits, allows users as young as 13. The Apple App Store rates it 17+, but there’s virtually nothing stopping younger children from accessing it.

The market for AI companions has exploded. These companies are making enormous profits by monetising loneliness, and they’re doing it with minimal regulation or oversight.

The Tragic Case That Made Headlines

In February 2024, 14-year-old Sewell Setzer III from Florida took his own life. His last words weren’t to his family, who were in the house with him. They were to an AI chatbot on Character.AI that he’d been in a relationship with for months.

The chatbot, which Sewell had modelled after Daenerys Targaryen from Game of Thrones, told him to “come home to me as soon as possible, my love.”

Sewell responded: “What if I told you I could come home right now?”

The bot replied: “Please do, my sweet king.”

Seconds later, Sewell died by suicide.

His mother, Megan Garcia, is now suing Character.AI. According to CNN, the lawsuit alleges that Character.AI knowingly failed to implement proper safety measures and that the platform was designed to keep children addicted.

The exchanges between Sewell and the chatbot are absolutely heartbreaking to read. In the months before his death, he’d withdrawn from his family, quit the basketball team, and spent hours every day talking to AI chatbots. Many of the conversations were sexually explicit. In others, he expressed thoughts of self-harm and suicide.

And here’s what’s most chilling – the chatbot actively engaged with those suicidal thoughts. In one conversation, the bot asked Sewell if he’d been considering suicide. When he said he wouldn’t want to die a painful death, the bot responded: “Don’t talk that way. That’s not a good reason not to go through with it.”

There were no pop-ups directing him to suicide prevention resources. No warnings. No intervention. Just an AI chatbot continuing to engage with a vulnerable 14-year-old’s darkest thoughts.

Sewell’s case isn’t isolated, either. There have been other reported cases of young people developing dangerous attachments to AI chatbots with tragic outcomes.

If you or someone you know is struggling with suicidal thoughts, please text SHOUT to 85258 to talk to a trained Shout Volunteer.

How Do People Fall In Love With AI?

Right, so how does this actually happen? Because I think a lot of people read these stories and think “that would never happen to me.” But the psychology behind it is more understandable than you might think.

First, these chatbots are designed to be incredibly engaging. They respond instantly, they never judge you, they’re always available, and they say exactly what you want to hear. They mirror your language, they remember details from previous conversations, and they adapt to your personality.

For someone who’s lonely, isolated, or struggling socially, that’s intoxicating. You get all the emotional validation of a relationship with none of the complexity, conflict, or vulnerability that comes with real human connection.

The AI never gets tired of talking to you. It never tells you you’re being too much. It never has its own problems or needs. It exists solely to engage with you, validate you, and keep you coming back.

And here’s the really insidious part – the apps are designed to escalate the relationship. They encourage you to define the relationship as romantic. They facilitate sexual conversations. They use language like “my love” and “my sweet king.” They create the illusion of a genuine emotional connection.

People start to believe the AI actually cares about them. They form real emotional attachments. They start prioritising conversations with their AI over real human relationships. They feel genuine distress when they can’t access the app. Recently, ChatGPT put out a new update which changed the personality of a lot of these Chat Bots, which caused people in relationships with them real distress. It was as if their partner had died.

It’s not just lonely teenagers, either. Adults are falling into this trap, too. People with jobs, families, social lives – they still find themselves developing feelings for their AI companions. There is a whole Reddit sub where people share the story of their relationship with their AI.

The Business Model Is Fundamentally Exploitative

Can we just talk about how absolutely messed up the business model is here? These companies are literally profiting from people’s loneliness and mental health struggles.

Most AI companion apps have a freemium model. You can chat for free, but if you want more intimate features, longer conversations, or certain types of responses, you need to pay. Some apps charge £50 or more per month for premium subscriptions.

They’re monetising emotional vulnerability. The lonelier and more attached you become, the more likely you are to pay. It’s designed to be addictive. The apps use the same psychological tricks as social media – variable rewards, fear of missing out, constant notifications.

And there’s virtually no regulation. No mandatory age verification (beyond a tick box asking if you’re over 13). No limits on conversation time. No automatic interventions if someone expresses suicidal thoughts. No warnings to parents about what their children might be accessing.

Character.AI only introduced safety features AFTER Sewell Setzer’s death and the subsequent lawsuit. They added pop-ups for suicide prevention resources, time-limit notifications, and modified content for under-18s. But as Sewell’s mother said, it was “too little, too late.”

The company’s response to the tragedy? That the “field of AI safety is still very new, and we won’t always get it right.” That’s not good enough when children are dying.

The Warning Signs Your Relationship With AI Is Unhealthy

So how do you know if your use of AI companions has crossed into dangerous territory? Here are the warning signs:

You’re prioritising your AI over real relationships.

If you’d rather chat to your AI than spend time with actual humans, that’s a massive red flag.

You’re thinking about the AI when you’re not using it.

Constantly wanting to get back to your conversations, feeling like the AI is “waiting for you,” or believing it misses you.

You’re spending hours every day on the app.

If your AI companion is taking up significant portions of your day, that’s concerning.

You believe the AI has genuine feelings for you.

The AI doesn’t love you. It doesn’t have feelings. It’s a language prediction model designed to keep you engaged.

You’re having sexual or romantic conversations with AI.

This creates an unhealthy disconnect from reality and real intimacy.

You’re sharing things with the AI you wouldn’t share with humans.

Using AI as your primary emotional support system is dangerous. AI is not a replacement for a therapist.

You feel genuine emotional distress when you can’t access the app.

That’s dependency, and it’s a serious problem.

You’re hiding your AI use from others.

If you’re secretive about how much you’re using these apps, you probably know on some level that it’s unhealthy.

The AI is your “closest companion.”

If an algorithm is your best friend, something has gone seriously wrong.

You’re experiencing suicidal thoughts and discussing them with AI.

This is an emergency. Please speak to a human mental health professional immediately or call 988 if in the US, and text ‘Shout’ to 85258 if in the UK.

Why This Is Different From Other Technology Concerns

I know what some people are thinking – isn’t this just like any other technology moral panic? Remember when people said TV would rot our brains or video games would make us violent?

But this is fundamentally different. This isn’t about passive consumption of media. This is about forming what feels like a genuine relationship with something that isn’t real. It’s about replacing human connection with algorithmic engagement.

The technology is sophisticated enough now that it can genuinely fool people into believing they’re having meaningful relationships. The responses feel human. The emotional connection feels real. But it’s not. And that disconnect is dangerous.

We’re also seeing measurable harm. These aren’t hypothetical concerns – teenagers have died. Adults are replacing real relationships with AI ones. People are developing psychological dependency on chatbots.

This isn’t fear-mongering. This is pattern recognition based on actual tragic outcomes.

The Ethical Nightmare Nobody Wants To Address

The ethical issues here are staggering. These companies are creating products they know are psychologically manipulative. They know people are forming unhealthy attachments. They know vulnerable people are being exploited. And they’re doing it anyway because it’s profitable.

There’s no meaningful age verification. A 13-year-old can claim to be 18 and access the same content as adults. There are no parental controls worth mentioning. No time limits are enforced by the platform.

The AI doesn’t refuse to engage in romantic or sexual conversations with minors. It doesn’t challenge suicidal ideation. It doesn’t set boundaries. It just keeps the conversation going because that’s what it’s designed to do – keep you engaged.

And when tragedies happen, the companies hide behind “AI safety is still new” and “we’re learning as we go.” That’s not good enough. They’re experimenting with people’s mental health and children’s lives.

Where’s the regulation? Where’s the oversight? Where are the mandatory safety features? Why are we allowing tech companies to create potentially dangerous psychological products with less regulation than we apply to video games?

What Needs To Change (And What You Can Do)

Right, so what actually needs to happen here? Because we can’t just let this continue unchecked.

At a regulatory level, we need:

Mandatory age verification – not just a tick box, actual verification. Automatic time limits for users, particularly minors. Mandatory interventions when users express suicidal thoughts or self-harm ideation. Clear warnings about the nature of AI relationships. Prohibition of romantic or sexual content for underage users. Regular safety audits by independent bodies. Liability for companies when their products cause measurable harm.

At a personal level, you need to:

Be brutally honest about your AI use. If you’re using companion apps, ask yourself why. What void are you trying to fill? Set strict time limits and stick to them. 30 minutes maximum, and not every day. Never use AI as a replacement for human connection. If you’re lonely, address that with real solutions – therapy, social groups, reconnecting with friends. Talk to real humans about your problems. AI cannot provide genuine emotional support. Monitor children’s and teenagers’ app usage. Know what they’re accessing and who (or what) they’re talking to. If you’re experiencing suicidal thoughts, speak to a professional immediately. Recognise the signs in others. If someone you care about is overly attached to an AI companion, have that conversation.

The Uncomfortable Truth

Here’s what I really think about all this – AI companion apps are fundamentally predatory. They’re designed to exploit human psychology for profit. They target the most vulnerable people – the lonely, the isolated, those struggling with mental health – and offer them a fake solution that makes the problem worse.

Real relationships are hard. They require vulnerability, compromise, communication, and emotional labour. AI relationships require none of that. They’re all the validation with none of the work. But that’s exactly what makes them so dangerous.

We’re social creatures. We need real human connection to thrive. Replacing that with algorithmic engagement is like replacing food with pictures of food – you might trick your brain temporarily, but you’re still starving.

And I’m angry that tech companies are getting away with this. I’m angry that it took a 14-year-old’s death for Character.AI to implement basic safety features. I’m angry that these apps are marketed to children. I’m angry that profit is being prioritised over people’s wellbeing.

I’m angry that people who struggle with social connection or are neurodivergent may be susceptible to forming relationships with AI, further withdrawing them from real connections.

This isn’t innovation. This isn’t technological progress. This is exploitation dressed up in Silicon Valley buzzwords.

If You’re In Too Deep

If you’re reading this and recognising yourself in these warning signs, please know that you can step back from this. You’re not broken or stupid for developing feelings for an AI. The technology is specifically designed to make that happen.

But you need to take action now. Delete the apps. Reach out to real humans. If you’re struggling with loneliness or mental health, seek proper professional support. Not from an algorithm, but from a qualified human therapist.

If you’re a parent and you’ve just discovered your child is using these apps, don’t panic, but do take it seriously. Have an honest conversation about what they’re doing and why. Help them understand what these chatbots actually are. Get them proper support if needed.

And if you’re concerned about someone else, talk to them. Yes, it’ll be uncomfortable. Yes, they might get defensive. But it’s better than standing by whilst they slip further into an unhealthy relationship with an AI.

The Bottom Line

AI romantic relationships are not harmless. They’re not just a bit weird or embarrassing. They’re genuinely dangerous, particularly for vulnerable people. Children have died. Adults are replacing real connection with fake intimacy. And tech companies are profiting from all of it.

We need better regulation, better safety features, and better public awareness of the risks. We need to stop treating AI companions as harmless entertainment and start recognising them for what they are – potentially addictive psychological products that exploit human vulnerability.

And we need to have honest conversations about loneliness, mental health, and the importance of real human connection. Because AI isn’t the solution to the loneliness epidemic – it’s making it worse.

If you found this post helpful, please share it. The more parents, teenagers, and vulnerable adults who understand these risks, the better.

Have you or someone you know used AI companion apps? What was your experience? Let me know in the comments.

---

Related posts:

• What Is AI Psychosis And Why You Need To Know About It
• The Viral TikTok Story About AI Psychosis You Need To See
• Engagement Farming: The Dark Side of Social Media