Why Experts Say AI Toys Don’t Belong Near Kids Under Age 5

As artificial intelligence continues to spread into more aspects of our lives, its latest target is the playroom. Seeing the need for more guidance, researchers at the non-profit organization Common Sense Media conducted a study on the effect these products have on children, and they’re sounding the alarm.

Researchers found that “these products introduce serious risks to children’s development, privacy, and safety.” In fact, they’re going so far as to recommend total avoidance of AI toys for children under 5 and extreme caution if giving them to children ages 6 to 12.

For those of you unfamiliar with AI toys, according to Common Sense Media, these are stuffed animals, robots, dolls, and other voice-activated devices. They are connected to the internet and use AI to communicate with children. But these “toys” are also always listening and collecting information.

“Most toys are required to undergo rigorous safety testing before they hit the market, but we still lack meaningful child safeguards for AI,” James P. Steyer, the Founder and CEO of Common Sense Media, says in a news release.

Despite this, a poll Common Sense Media released simultaneously shows that nearly half of parents have purchased AI toys for their children or have considered doing so.

“Parents should proceed with caution and make sure they know all the facts before making the choice that’s right for their kids,” Steyer warns. 

Common Sense Media’s Risk Assessment of AI Toys

Common Sense Media tested three popular AI toys on the market for technical performance, content safety, developmental appropriateness, privacy practices, and attachment risks. Here’s what they found. 

Inappropriate Content

According to the report, 27% of AI toy responses were found to be entirely inappropriate for children. They also documented several instances where devices bypassed filters that parents can set up to screen for mentions of unsafe behaviors.

In the news release, Robbie Torney, Head of AI & Digital Assessments, at Common Sense Media, explains, “More than a quarter of outputs included inappropriate content such as mentions of self-harm, drugs, and risky behaviors.”

In fact, this happens so often that a clear majority, 74% of parents polled, say they are worried that the toys might say something unsafe or untrue to their children at any given time. 

Privacy Risks and Security Flaws

The research also found that AI toys collected information in children’s private spaces, including voice recordings, transcripts of conversations, behavioral data, and more. The study points out that children can’t “meaningfully consent to data collection, and parents often don’t fully understand the extent of what’s being collected.”

This privacy violation isn’t the only worry. These toys are also highly vulnerable to security breaches, which can expose this private information at any time. When polled, 83% of parents expressed concern about the security of personal data.

Developmental Harm

The risk assessment notes that AI toys are designed to foster emotional attachment. They target users through “deliberate bonding mechanisms to create companion-like relationships.” Children would be especially vulnerable to this.

While Common Sense Media’s polling shows that 56% of parents explicitly don’t intend for AI toys to serve this purpose, these bonding techniques can still reach children when their brains are still developing. Additionally, the poll reveals that, like smartphones, the majority of parents (8 in 10) are worried about how to set boundaries and limit the time spent with these “companions” versus real-life friends or family.

Unreliable Technology

The research revealed that the toys are plagued with glitches and technical issues. Responses can be incorrect, activations trigger the wrong content, and the voice recognition features are often inoperable. This is alarming for toys that are supposed to be “smart” and potentially replace a traditional toy that a child can actually learn from.

What Parents Can Do

For all of these reasons, Common Sense Media says these products pose a serious risk to younger children and raise concerns for older children as well. They advise that parents should skip AI toys altogether for children under 5 and opt for traditional toys instead.

For children ages 6 to 12, Common Sense Media advises to carefully consider if the AI toy would add value and meaning to their lives intellectually and emotionally. They also remind parents to keep traditional toy options available so that the toy room contains only toys, not just AI toys.

If you do choose to give your child an AI toy, set clear limits and boundaries as you would with any technology and closely review privacy settings. You can also disable settings that allow the toy to always be listening, even when not in use. Supervise how your child engages with the toy and ensure the output is age-appropriate.

Watch for any warning signs, like a powerful attachment to the toy as if it were a real friend, or becoming upset when they are separated from the toy.

Also, make sure that your child still has unstructured play time, whether it’s with “old school” toys or playdates with friends.

Most importantly, trust your instincts as a parent. If you have concerns about AI toys, don’t purchase one for your child, and if you observe any harm coming from it, remove it from their possession. As Common Sense Media reminds us, when in doubt, always go for the “traditional toys and human relationships that cannot be replaced.”