AI Toy Security Flaw Exposed Over 50,000 Kids' Chat Logs to Anyone With a Gmail Account

Security Flaw Leaves Children's Data Wide Open

When security researcher Joseph Thacker's neighbor asked him about Bondu, an AI toy she'd pre-ordered for her children, he decided to investigate. What he and fellow researcher Joel Margolis discovered in just minutes was alarming: the company's web portal vulnerability left children's private conversations completely exposed to anyone with a Gmail account1

1

2

2

. Without any hacking required, the researchers simply logged into Bondu's public-facing web console using an arbitrary Google account and immediately gained access to transcripts of virtually every conversation the AI-powered toys for children had ever conducted.

Over 50,000 Chat Transcripts Accessible Without Protection

The data exposure was staggering in scope. Bondu confirmed to the researchers that more than 50,000 chat transcripts were accessible through the exposed web portal, representing essentially all conversations the toys had engaged in except those manually deleted by parents or staff2

2

3

3

. The sensitive personal information left unprotected included children's names, birthdates, family member names, parental "objectives" for the child, and detailed summaries of every chat between each child and their Bondu. The researchers also saw pet names kids had given their toys, their likes and dislikes, favorite snacks, and dance moves—intimate details shared with what children believed was a trusted companion.

Authentication Measures Implemented After Alert

"It felt pretty intrusive and really weird to know these things," Thacker told WIRED. "Being able to see all these conversations was a massive violation of children's privacy"1

1

. When the researchers alerted Bondu to the glaring security flaw, the company took down the console within minutes and relaunched it the next day with proper authentication measures. Bondu CEO Fateen Anam Rafid stated that security fixes "were completed within hours, followed by a broader security review and the implementation of additional preventative measures for all users"2

2

. The company reported finding no evidence of access beyond the researchers involved and has since hired a security firm to validate its investigation and monitor systems going forward.

Privacy Risks With AI Toys Extend Beyond This Incident

While the immediate vulnerability has been addressed, the incident highlights broader data security challenges facing the AI toy industry. Margolis characterized the exposed information as "a kidnapper's dream," explaining that "we're talking about information that lets someone lure a child into a really dangerous situation, and it was essentially accessible to anybody"3

3

4

4

. The researchers' glimpse into Bondu's backend revealed how AI-powered toys maintain detailed histories of every chat to personalize future conversations—creating extensive data collection that poses ongoing risks even when properly secured. Margolis noted that "all it takes is one employee to have a bad password, and then we're back to the same place we started, where it's all exposed to the public internet"1

1

.

Questions About Third-Party Data Sharing

According to what the researchers observed in the admin console, Bondu appears to use Google Gemini and OpenAI's GPT models to power its conversational capabilities1

1

4

4

. This raises additional concerns about whether user data from children's conversations is being shared with these technology companies. The researchers suspect the web console might have been built using AI coding tools, which can generate functional-looking code riddled with security holes4

4

. Bondu did not respond to questions about whether AI built the infrastructure meant to protect children's privacy. While the company stored only written chat logs and auto-deleted audio recordings after short intervals, the incident serves as a stark reminder that parental monitoring tools cannot protect against vulnerabilities in the toy companies' own systems. Parents evaluating AI toys should consider not just device security features, but the data handling practices and security posture of manufacturers themselves.