Is AI Affecting Your Child’s Mental Health?

Care Alliance Counselling
Jul 3
5 min read

There is a good chance your child has already had a conversation with an AI chatbot in some capacity. Maybe to help them with their homework, or even create funny images to share with their friends. Nowadays, millions of young children’s and teens’ social media feeds are flooded with algorithms specifically designed to keep them scrolling for hours on end.

Artificial intelligence is no longer a distant concept, and it has taken over our everyday lives. As the AI space continues to grow, this post is not intended to alarm you or dismiss your concerns. It is here to give you an honest, research-grounded look at what we know, so you can navigate this with your child from a place of understanding rather than fear.

Two Different Faces of AI

When talking about AI and children’s mental health, we are really talking about two distinctly different areas that are often lumped together.

The first is social media algorithms. The invisible systems that have successfully determined what content children see on platforms such as TikTok, Instagram and YouTube. These are not passive tools and have been meticulously designed over the years to learn what type of content captures a user’s attention and then relentlessly deliver that content to their intended audience.

The second is AI chatbots. Generative AI tools such as ChatGPT, Character.AI, Replika and Claude have become so accessible and normalised that children can make a free account and simply talk and interact with these bots directly, forming incredibly emotional and strong connections without even realising the repercussions. When they have finished their free trial with one bot, they can create a new account with another and continue this pattern, constantly having an online companion to chat with.

Both these methods interact and manipulate children’s mental health, but in meaningfully different ways and it’s important to realise the distinction between them.

The Algorithm Problem: When the Feed Becomes a Mirror

Children who spend more than three hours per day on social media face twice the risk of developing poor mental health, including increased anxiety and depression (First Focus on Children, 2024). While this is a striking figure, it is important to understand why this happens.

The answer may lie less in social media itself and more in how it is designed. Social media algorithms are built to promote whatever content a user is interested in. So, if a teen searches for mental health content (e.g. depression, anxiety or self-harm), their algorithm will continue to show them more of that particular type of content, creating a harmful cycle where they begin to believe that everyone around them feels the same way and is also seeing the same content (Yale Medicine, 2024). Compulsive engagement with these platforms has been associated with increased symptoms of anxiety, depression and attention disorders, with algorithm-driven content personalisation that has contributed to prolonged use, which has reinforced addictive behaviours similar to those seen in individuals with substance use disorders (PMC, 2025).

This is particularly disturbing for perceptive teens who are already struggling with mental health issues. Research points to heightened risks surrounding body image, disordered eating, cyberbullying and disrupted sleep, especially concerning teenage girls, further amplified by their social media algorithm content (Ng, 2025).

What this means for younger children (under 10): They are not yet on social media platforms; however, this does not mean that they are immune to the internet. Algorithm-driven content on YouTube and similar platforms used for educational purposes can still shape what a young child sees and normalises, often without their parents realising it.

What this means for teenagers: This age group is especially vulnerable because adolescence is precisely when their identities, peer pressure, and the need for social validation are the most intense. Each social media platform’s algorithm is not neutral and is solely optimised for the most efficient engagement, not a teenager’s wellbeing.

The Chatbot Question: Connection or Concern?

Here is where the picture becomes more nuanced than a simple fix of making sure to proactively monitor your child’s screen time.

A 2025 RAND survey of 1,058 adolescents and young adults aged 12-21 found that roughly one in eight were already turning to AI chatbots for mental health advice, with two-thirds engaging with them at least monthly and more than 93% reporting that the advice from the chatbots was helpful (RAND, 2025). Researchers note that the high utilisation likely reflects the low cost, immediacy and perceived privacy of using AI, which is particularly appealing to young individuals who face barriers to traditional counselling methods (RAND, 2025).

Traditional mental health services face severe accessibility challenges, with waiting times for child psychiatric consultations ranging from months to years in certain regions (PMC, 2025). For a teenager in distress at 11 PM who cannot access a therapist, their best option is a chatbot that responds instantly. Research does suggest some benefits to this “secondary therapist”: a systematic review and meta-analysis found that chatbot-delivered interventions/therapy significantly reduced psychological distress in young people, though the broader effects on wellbeing were less conclusive (Shi et al., 2025).

But before you cancel your weekly therapy session and turn to ChatGPT, the risks deserve honest attention too. Reports indicate that 72% of teens have turned to AI companions and 33% have developed emotional relationships with these chatbots (PMC, 2025). When a child develops a deep emotional bond to an AI, one that is readily available 24/7 and still learning how to make its own decisive decisions, it erodes the developmentally complex work of building real human relationships that we all need and rely on in our everyday lives.

More importantly, the adolescent mental health landscape is rapidly evolving as AI reshapes how young people connect and navigate developmental challenges, from the known risks of algorithms reinforcing pathology on social media to the emerging risks of generative AI fostering emotional attachments in vulnerable youth (Ng, 2025).

What the Research Cannot Tell Us (Yet)

As discussed previously, the field of AI is constantly evolving, and there is so much that we have yet to uncover. Only 16% of the Large Language Model (LLM) chatbot studies have undergone clinical efficacy testing, with most still in early validation stages (Tavabi et al., 2025). We are, in many ways, running an uncontrolled experiment on an entire generation of children with unpredictable outcomes.

This does not mean panic is warranted; it means thoughtful, informed attention is.

What This Means for You

For parents and caregivers:

You do not need to ban screens or become an expert in AI. But there are a few things that are worth taking note of:

Stay curious, not interrogative: Ask your child what they use, who or what they spend their time talking to online. An algorithm feeding a struggling teenager dark content for one hour is more harmful than two hours of benign content.
Treat AI chatbots as a conversation starter, not a crisis: If your child is turning to chatbots for emotional support/comfort, this is useful information. Do not shut down this relationship, ask them what they gain from this situation. Their response will tell you what they need from you or perhaps a professional.
Maintain the primacy of human connection: Research consistently shows that strong, secure relationships with caregivers are the most protective factor for children’s mental health (Ng, 2025). No algorithm or chatbot can change that. Your presence, attention and willingness to sit with your child through difficult times trumps any technology.

The Takeaway

AI and technology are vital tools that are not going away any time soon. The goal is not to shield your children from them, but to help them navigate through these challenges with enough self-awareness, connection and support so that the technology serves them - rather than the other way around.

At Care Alliance Counselling, we work with children, adolescents and families navigating exactly these kinds of pressures. If you are concerned about your child’s relationship with screens, social media or technology, or if you are noticing changes in their mood, sleep or social behaviour, we are here to help.

Kayla Pereira - Summer Intern 2026