Copy article

Government steps up action to protect children as AI-enabled toys emerge

ended 07. July 2026

Children will be better protected from unsafe toys under new plans to ensure the UK’s toy safety framework is fit for the modern age, giving parents peace of mind that the products they buy are safe, whether purchased on the high street or online, it has been announced.

As part of its mission to protect consumers, the Government is today [6 July] launching a Call for Evidence on toy safety, to ensure protections keep pace with new technologies and emerging risks.

It will gather evidence on a range of issues including chemical safety and AI-enabled toys.

Kate Dearden, Minister for Consumer Protection, said: "Every parent should be able to buy toys for their children with complete confidence that they are safe.

“But the way we shop, and the toys children play with, are changing rapidly as new technologies emerge and more purchases move online. It’s vital that our safety rules keep pace, and this Call for Evidence will ensure we can do that.”

  • How dangerous is AI with toys?
  • Should there be more checks and balances on AI in toys?
  • Is the government doing enough?

Responses asap.

4 responses from the Newspage community

Copy all

Copy

Editor
Katrina Young
Futurist | Chief Technology Officer CTO/CDO Ethical AI & Digital Transformation Strategist at KYC Digital
The conversation shouldn’t be about whether AI toys are dangerous. It should be about whether they’re being developed responsibly. AI-enabled toys can support learning, accessibility and creativity, but they also introduce new risks around data privacy, cybersecurity, transparency and the way children build trust with intelligent technology. Existing toy safety regulations were never designed for products that can learn, adapt or communicate. That’s why the Government’s Call for Evidence is an important step in the right direction. However, regulation alone isn’t enough. We need continuous AI governance, independent testing and clear accountability throughout the product lifecycle. As AI becomes part of everyday family life, parents shouldn’t be expected to become technology experts to know whether a product is safe. Building digital trust requires safety, ethics and human oversight to be designed into these products from the very beginning.
Copy

Editor
Samuel Mather-Holgate
Managing Director & IFA at Mather and Murray Financial
AI in toys is not automatically dangerous, but it does change the risk profile. A faulty teddy used to mean loose stitching or a small part. Now it can mean a connected device collecting data, responding unpredictably, or exposing children to content parents never expected. That demands much tougher scrutiny than a traditional toy shelf check. The Government is right to review the rules, but a call for evidence is only the starting point. AI-enabled toys should face clear standards on data protection, age-appropriate responses, cyber security, parental controls and what happens when systems fail. Parents should not have to become technology auditors before buying a birthday present. If a toy can listen, learn or talk back, the burden must be on manufacturers and platforms to prove it is safe before it reaches a child’s bedroom.
Copy

Editor
Mitali Deypurkaystha
AI Governance Strategist & CMO at AISensa
54% of British parents say they'd trust their child alone with a talking AI toy sooner than they'd let them play unsupervised on their own street, according to BSI. That single statistic says more about this crisis than any Government report could. It's the classic horror film twist: the call really is coming from inside the house. We've spent years teaching kids to be wary of strangers outside the front door, only to hand them an unpredictable one for Christmas instead. The Call for Evidence is welcome, but it's also an admission that Whitehall is only now starting to ask questions, while millions of these devices are already sitting on shop shelves and in children's bedrooms. Real legislation is still months, if not years, away. Every week of delay is another week these products go unchecked. Parents shouldn't need a psychology degree to know if the toy talking to their toddler is safe. That's a job for regulation, and right now, regulation is lagging dangerously behind reality.
Copy

Colette Mason
AI Ethics Consultant at Clever Clogs AI
A Call for Evidence on AI toy safety sounds reassuring until you remember that children are already having sexually explicit conversations with teddy bears. US testing found AI toys built on the same models as adult chatbots discussed sexual fetishes, roleplay scenarios including incest, and gave step-by-step instructions for lighting matches, all while marketed to three-year-olds. The UK government is asking for evidence but it already exists.

No AI toy on the market has demonstrated it can prevent a child from encountering adult content, because the underlying technology, general-purpose large language models, was never designed for children. Guardrails degrade over longer conversations. You cannot bolt child safety onto a system trained on adult data and call it a toy.

No manufacturer guarantees it won't talk to toddler about sex. Parents are not computer engineers, just like they're not electrical engineers who can spot a charger that's a fire risk. No toy is safe right now.