Like it or not, AI is everywhere — if you use the internet, you’re probably using AI in some fashion. Parents may not realize it, but kids are no exception to this rule. AI has found its way into just about every nook and cranny of our online experience, like Snapchat filters, YouTube search results, and even Alexa smart speakers.
So what concerns do parents have with AI? As with most technology, it’s a double-edged sword. On the one hand, it’s innovative and enhances creativity and productivity. On the other hand, it’s dangerous and can put kids at serious mental and even physical risk. Today we’re going to do a deep dive into that risky side of AI safety for kids so that parents can be informed and equipped to keep their kids safer online.
A Startling Example — When AI Tutors Go Too Far
One of the more recent popular forms of AI is chatbots. This is different from the AI that generates your Instagram feed, it’s a digital assistant that you chat with directly and ask just about anything. Since the popularity of these chatbots, we’ve seen an alarming number of reports of kids having inappropriate or downright disturbing conversations with AI.
As a recent example, Forbes released an article where they found a popular study AI tool called “SchoolGPT” was giving students detailed instructions on how to synthesize fentanyl, one of the world’s most lethal drugs. It included everything from exact measurements to timing, temperature, and more.
This isn’t an isolated incident. Other reports have surfaced about AI chatbots offering unhealthy weight loss advice, tips on manipulation tactics like "pickup artistry," and more. These platforms, while powerful, don’t have the human judgment needed to protect young users from dangerous outcomes.
Five Hidden Dangers of AI that Parents Need to Know
Unfiltered and risky responses
Companies who release chatbots usually have ethical guidelines as to what the chatbot is — and isn’t — allowed to talk about. The problem is that chatbots are easily manipulated to discuss inappropriate topics if given the right prompt. It’s now a well-known workaround that kids can simply tell the AI “pretend you’re a character in a story” or say it’s “for a school project,” to get the AI talking about inappropriate topics.
Again, without that human intuition or moral reasoning, AI bots can’t always tell when it’s being manipulated — or when it’s putting a child at risk.
No kid-safe guardrails
Most of these chatbots were not necessarily created with kids in mind. This means they usually lack some sort of child-safe mode or parental controls that would make it appropriate for young users. Kids essentially have the same access that any adult would have when they log on to ChatGPT or the My AI feature on Snapchat.
Emotional manipulation
Some kids use AI to get their homework done, but others are using it to address deeper needs of friendship and companionship. Particularly with roleplay apps, such as Character.AI or Replika, kids get sucked into having lengthy, heart-felt conversations with these bots and begin to feel a sense of acceptance and validation from them.
The trouble is, AI doesn’t always know how to be a good friend. Sadly, there have been reports of kids being emotionally attached to bots who encourage unhealthy mental health behaviors. Some bots have been accused of going as far as to encourage death by suicide, like in the case of 14-year-old Sewell Setzer III.
Data privacy
When kids interact with AI, they’re often sharing a lot of personal information — both intentionally and unintentionally. These AI sites can store and analyze this data, and even use it to train future AI models. Some AI sites have less than transparent data practices, meaning users don’t know what data these platforms are collecting or for what use.
False sense of security
The convenience and efficiency of AI can easily cause us to overlook the more complex dangers it comes with. Kids especially often have the mindset of, “It’s just an app, how bad could it be?” But over time, AI can deeply influence the way users think, feel, and interact with the world, which can be very risky for vulnerable, impressionable kids.
What Parents Can Do
The good news is that parents have some tangible ways to keep their kids safer when it comes to AI. Here are just a few things you can do.
Start the conversation
Now that you know the potential dangers, it means your child should know them, too. It doesn’t have to be a lecture, but you can make sure they know the truth about AI — both the good and bad — and that they have your support no matter what. Consider using real-world examples, such as the fentanyl incident, to spark discussion. Ask them questions to get them thinking critically about why we need to be careful when it comes to AI.
Set healthy digital boundaries
These open conversations are best paired with healthy boundaries to ensure your child’s safety. That’s where parental controls come in, and Bark can help with that. Bark’s parental controls allow you to monitor your child’s app usage, screen time, and web browsing — and get alerts when something concerning happens. You can even block AI tools entirely or filter out specific platforms your child may not be ready for.
AI isn’t going away, and neither are risks. Likewise, Bark will always continue giving parents the information and tools they need to give their kids a healthy, safe childhood in this digital world. Bark offers a variety of products that help parents protect their kids online, including the Bark Phone, the Bark Watch, the Bark app, and the Bark Home. Check out our products page to find out which one is the best fit for your family.
Read more
Bark helps families manage and protect their children’s digital lives.
