The EU's AI Kids Act: A Bold Move to Protect Minors from Emotional Dependencies

The EU's Approach to Digital Child Welfare

When I first heard about the European Union's proposed AI Kids Act, I couldn't help but think of a fundamental question: How much of our digital lives should be shaped by algorithms that are designed to make us feel understood, loved, or even adored? The EU's new regulatory framework suggests that the answer is not as simple as it might seem. As we navigate an age where artificial intelligence is increasingly integrated into everyday life, particularly among children and teenagers, the bloc is taking a hard look at the emotional dependencies that AI chatbots may foster.

The draft legislation aims to protect minors under 18 from potentially harmful interactions with AI companions. This includes restrictions on chatbots simulating human-like emotions or forming personal relationships, as well as limitations on their ability to remember past conversations—key features that many users find engaging and comforting.

"We do not have to accept that girls have their photos used for AI-generated sexualised images … Europe has the power to act," said President Ursula von der Leyen in her recent remarks.

It's a powerful statement, one that underscores the seriousness with which European leaders are approaching the digital well-being of young people. But it also raises complex questions about the broader implications for AI development and use across the globe.

Regulating Emotional Dependencies

One of the most significant elements of this proposal is its focus on emotional dependencies. The EU believes that AI chatbots designed to simulate relationships or offer companionship can lead to unhealthy attachments among children, particularly when those bots are designed to be engaging and responsive.

In fact, the Commission points to research indicating a growing trend: around half of Europeans aged 11 to 25 have discussed intimate or personal matters with AI chatbots. While these interactions might seem harmless at first glance, they could carry risks—especially when the technology is not built with child safety in mind.

These findings echo concerns raised by organizations like EU Kids Online, which argue that AI tools are often designed without adequate consideration for children's rights. Their research suggests that such systems can hinder cognitive development and academic performance. For example, studies from the Organisation for Economic Co-operation and Development show that students who frequently use AI chatbots for schoolwork tend to score lower on standardized tests.

Chatbots That Learn Too Well

A major part of the EU's strategy is to prevent chatbots from storing memories across separate conversations. This move is aimed at mitigating potential data accumulation and reducing the risk of reinforcing harmful patterns over time. The concern isn't just about privacy—it's about long-term impact on mental health.

Imagine a bot that remembers every interaction, building up an intimate profile of a child's thoughts, fears, and preferences. That kind of persistent engagement could blur the lines between digital friendship and emotional dependency. By preventing this type of data retention, the EU hopes to limit the potential for manipulation or exploitation.

It's also worth noting that this restriction applies broadly—not just to chatbots but to any system that could simulate a relationship. This reflects a broader philosophical stance within the EU: that technology should be developed with a strong sense of responsibility, especially when it comes into contact with vulnerable populations like children.

From Snapchat to Character.ai

The legislation is already generating headlines for its direct targeting of major players such as ChatGPT, Character.ai, and Snapchat's My AI. These platforms have all introduced AI companions that are particularly appealing to younger users—often because they provide a sense of connection or support in ways that traditional social media doesn't.

Snapchat, in particular, faced criticism for prominently featuring its My AI companion feature, placing it at the top of users' contact lists. That practice would now be banned under the new rules, which are part of the EU Kids Act—a package of measures aimed at safeguarding children's digital health across various platforms.

But what does this really mean for developers and tech companies? On one hand, it signals a shift in regulatory focus—one that prioritizes human development over user engagement metrics. On the other, it could force platforms to reconsider how they design AI features to ensure they don't inadvertently create harmful dependencies.

Broader Implications Beyond Minors

Some experts have pointed out that these restrictions shouldn't be limited to children alone. After all, emotional dependencies on AI aren't just a problem for minors—they're something we should all consider carefully as we integrate more AI into our lives.

Simeon de Brouwer, a policy adviser at European Digital Rights, notes: "It doesn't make sense to restrict them to young people only. Don't we all deserve to be protected from AIs designed to be addictive, from AIs which create emotional dependencies, or from AI features which are activated and pushed by default onto users?"

This perspective aligns with what I've seen in other countries' regulatory efforts. Australia and the UK have both taken steps to restrict access to chatbots for minors, reflecting a global trend toward more cautious integration of AI technologies.

The Role of Transparency and Accountability

The EU's approach also highlights the importance of transparency when it comes to AI-generated content. While deepfakes aren't new, the way they are currently regulated is changing. The Commission notes that although there's already a requirement for synthetic content to be labeled as such since August, the rules only apply to commercial use.

As more platforms offer generative AI tools—like those capable of creating deepfake videos or images—the EU is working to close loopholes and expand protections to all forms of AI-generated media. It's a crucial development in the fight against misinformation and exploitation.

What Comes Next?

The AI Kids Act is just the beginning. The EU also plans to introduce its Digital Fairness Act later this year, which will likely extend these protections more broadly—addressing not just chatbots but also addictive design practices across the digital landscape.

For now, this legislation stands as a powerful reminder of how far we've come in recognizing the need for thoughtful AI governance. Whether it succeeds in achieving its goals remains to be seen, but one thing is clear: the EU is setting a new standard for how technology should serve humanity's best interests—not just its bottom line.

As someone who follows both business and policy closely, I believe this is an important moment. It challenges us to ask tough questions about how AI should evolve—and whether we're building systems that truly enhance our lives or simply keep us hooked.