The ICO’s provisional findings reveal shortcomings in Snap’s assessment of privacy risks associated with AI.
The UK’s data protection regulator is scrutinizing Snap’s AI chatbot, expressing concerns about potential privacy risks to children. The Information Commissioner’s Office (ICO) announced today that it has issued a preliminary enforcement notice to Snap, citing a “potential failure by Snap to adequately assess the privacy risks posed by its generative AI chatbot ‘My AI’.”
Importantly, it’s worth noting that the ICO’s action doesn’t constitute a finding of a breach but rather signals that the regulator has reservations about Snap’s efforts to ensure compliance with data protection rules, including the Children’s Design Code implemented in 2021.
The ICO’s investigation has provisionally found that Snap’s risk assessment before launching ‘My AI’ inadequately addressed the data protection risks associated with generative AI technology, especially concerning children aged 13 to 17.
Snap now has the opportunity to respond to the regulator’s concerns before the ICO makes a final determination regarding potential rule violations.
John Edwards, the Information Commissioner, commented on the provisional findings, stating, “The provisional findings of our investigation suggest a concerning failure by Snap to adequately identify and assess the privacy risks to children and other users before launching ‘My AI.’ We have been clear that organizations must consider the risks associated with AI, alongside the benefits. Today’s preliminary enforcement notice shows we will take action to protect the privacy rights of UK consumers.”
Snap introduced the generative AI chatbot in February, with its UK debut in April, using OpenAI’s ChatGPT large language model technology. The chatbot was designed to act as a virtual friend, offering advice and receiving snaps from users.
Initially available to Snapchat+ subscribers, Snap eventually made “My AI” accessible to free users and added the capability for the AI to send snaps to users who interacted with it, generated using generative AI.
While Snap claimed to have implemented additional moderation and safety features, including default age consideration, to ensure user-appropriate content, reports emerged of the chatbot providing inappropriate recommendations.
Related stories:
