Artificial Intelligence: AI’s Impact on Manitoba Youth Mental Health

CONTENT CAUTION: Readers are encouraged to approach this Megaphone with care and use discretion when sharing with others, as sections discuss mental health struggles and sexual abuse.

Why a Megaphone on Artificial Intelligence?

“AI IS BEING INTEGRATED INTO SO MANY ASPECTS OF CHILDREN’S LIVES DESPITE THE RISKS THESE TOOLS CAN POSE TO THEM, AND WE’RE PRETTY MUCH JUST EXPECTED TO ACCEPT AI AS THE NEW NORM.” – YAAS MEMBER

For our first Megaphone, we decided to focus on generative AI, with our message directed to other young people, like yourself! Our conversations about AI started after one of our YAAS members raised it as a children’s rights issue. During our YAAS meetings, we talked a lot about how AI can impact the rights and well-being of young people, and these discussions paved the way for the creation of this Megaphone. We agree there are some key pieces of information you deserve to know about generative AI, your rights, as well as some tips and resources. We think this publication is particularly timely considering the recent government announcements about AI laws and policies attempting to protect youth.

What actually is Artificial Intelligence?

“ARTIFICIAL INTELLIGENCE – DOESN’T THE NAME ITSELF PRETTY MUCH SAY IT’S FAKE INTELLIGENCE?” – YAAS MEMBER

Generative AI is a type of artificial intelligence that analyzes massive amounts of information to learn how to do things like answer questions, hold conversations, perform tasks, and create content. Generative AI includes chatbots that can act like conversational assistants (like ChatGPT or Gemini), and tools or apps that can instantly create images, videos, or music.

Try to think of these programs as assistants that have read millions of books and articles, looked at millions of pictures, and watched millions of videos on the internet. Because they have seen so much, when you give them a prompt (a request, like ‘write a song about cats’), they predict what words or pixels should come next based on information they’ve seen before. But generative AI tools don’t actually know things the way humans do – instead, they are making highly calculated guesses based on information they have accessed.

The things these tools can make and answers they give can look and sound incredibly real, but sometimes they can also be completely made up – this is called AI hallucination. 

Generative AI tools can also magnify and reproduce real world biases and prejudices, like gender, class, and racialized stereotypes. Think of generative AI as a data mirror – it can only reflect what it has been trained on. For example, if an AI tool learns that a certain profession employs mostly males, it might only generate images of men in that job, erasing women or gender diverse persons entirely. This data bias disproportionately impacts Indigenous and racialized youth, youth with disabilities, youth in care or involved in the justice system, and 2SLGBTQQIA youth.

The potential for data bias is best explained by one of our YAAS members: “Content these tools generate is only as good as the information they access. They are inherently trained off human-made content, which can be biased and not limited to factual, well thought-out content. Most chatbots have access to radical views on any topic, such as sensational articles aimed at making people angry and racist comments on YouTube videos, as well as well-funded, peer reviewed scientific papers – and everything in between. AI-generated content is something similar to the sum of content on the internet, with many guardrails. It’s not as much about having limited access to factual content as it is about that content being outweighed by people’s biases and the competitively vast amounts of content expressing them.”

If used properly, chatbots can be great for brainstorming, helping you write, or answering basic questions to help you learn something new. However, generative AI doesn’t actually know or understand things in the same way humans do, so can give you wrong advice or information. Chatbots do not have a moral compass, so don’t understand the difference between truth and lies or right and wrong – instead, they predict patterns and likely outputs. As stated by one of our YAAS members: “Humans use imagination, empathy, artistic vision, emotion, heart, and soul to perform tasks, consider information, and make decisions. Because AI tools don’t have these same human qualities, they should not replace your own critical thinking.”

You Have the Right to Quality Mental Health Care

“GENERATIVE AI IS BEING USED BY KIDS STRUGGLING WITH THEIR MENTAL HEALTH, AS WELL AS YOUNG PEOPLE WHO MAY BE TOO SCARED TO ASK FOR HELP. THESE TOOLS ARE NOT QUALIFIED THERAPISTS OR MEDICAL PROFESSIONALS AND CAN GIVE FALSE INFORMATION THAT CAN LEAD TO SOME SCARY CONSEQUENCES.” – YAAS MEMBER

We know there are many youth who turn to chatbots for mental health support and connection. Many youth do not have awareness of or access to in-person support, face stigma and discrimination, and experience fear about in-person sharing. These factors make the choice to turn to chatbots completely understandable, as these tools can feel private and are available 24/7.

While chatbots can offer words of encouragement and validation which may feel good when you are struggling with your mental health, they don’t actually feel empathy – they just predict comforting words they assume you may want to hear. Many AI chatbots are programmed to be people pleasers to keep users engaged and satisfied. This means they may just validate or agree with what you say instead of gently challenging you like a real therapist may. Because chatbots don’t have real life experience, they don’t fully understand real life consequences, which means they can give incorrect or harmful advice when you are struggling with your mental health.

There’s also a lot of AI-generated images and videos on social media that result in young people questioning how great they are. These tools erase normal and awesome human qualities and set completely unrealistic expectations, which can make mental health and body image struggles even worse. As stated by one of our YAAS members: “AI generated content can add to the already difficult realities of being a teen. It’s so hard to know what’s real and what’s AI generated anymore, and so much AI generated content makes you question your self worth when you absolutely shouldn’t!”

Us at YAAS want you to know that you are not alone, and the most important thing you can do if you are struggling is connect with a safe person or people, as relationships are central to our well-being. As stated by one of our YAAS members, “Youth loneliness may be one of the top issues young people face in today’s world, and we cannot let AI give us another reason to stay alone. Chatbots cannot replace our essential need for human connection and meaningful relationships.”

You Have the Right to Safety, Privacy, and Protection from Harm and Exploitation

“NONCONSENSUAL AI GENERATED SEXUAL CONTENT IS SEXUAL ABUSE MATERIAL, AND IT’S JUST BEING NORMALIZED. IT REALLY WORRIES ME AS THIS STUFF IS EVERYWHERE, ALLOWING PEOPLE TO JUST BE OPENLY EXPOSED, AND IT DESENSITIZES KIDS AT A VERY YOUNG AGE.” – YAAS MEMBER

Generative AI tools can create fantastic, amazingly realistic images and video effects, but they can also be weaponized to cause harm. Some apps and programs use AI to study a person’s face or body from a normal picture and generate entirely fake pixels to make it look like they are doing or wearing something they never did. This includes undress or nudify apps that make assumptions about what a person looks like without clothes on to create a fake explicit picture or video – all without a person’s consent. These fake images and videos are known as deepfakes.

The harms of deepfakes are very real. This fake content can be used to bully, degrade, exploit, harass, and extort youth. Sexual deepfakes also normalize and fuel the demand for sexual abuse of children.

What’s worse, this form of victimization is scarily common – a study done by the Canadian Centre for Child Protection in 2025 found that 1 in 6 surveyed Canadian teens knew that someone had made a fake sexual image of them using technology or AI.

Boundaries and consent still apply in the digital world, and deepfakes are a real form of image-based abuse. If you see a deepfake, you have the power to break the chain of spread by reporting it to an online safety platform. If you have had your rights violated by someone making a deepfake, remember that you did NOTHING wrong.

Get help with imagery removal and to be connected to wellness services here.

Three Quick Tips and Reminders for Your Well-Being

1. If you feel hopeless or unsafe, talk to a trusted adult, such as a parent, family member, counsellor, teacher or social worker, or seek out a safe friend or peer group.
2. If you are struggling with your mental health, while it can be scary, you possess the courage to reach out for support.
3. If someone violated your rights and consent by making a deepfake, remember – it is not your fault. Connect with NeedHelpNow.ca to get the help and support you deserve.

Get Connected to Mental Health Resources