Meta adjusts AI after bot questioned woman about her daughters
What's the story
Meta has announced plans to change the prompts suggested by its AI chatbot. The decision comes after a viral video showed the bot prying for personal information about a woman's young daughters, as reported earlier by Futurism. In a statement to The Verge, Meta's spokesperson Dina El-Kassaby admitted that the company "missed the mark" with this feature, saying it "never should have prompted the individual with questions like that."
Incident details
What did the viral video show?
The controversy was sparked by Instagram user Kalie Robins, who shared a video of her and her child singing in the car.
After cross-posting the clip to Facebook, Meta AI suggested an invasive question: "Who's the child passenger?"
When Robins clicked on this prompt, the AI pulled information about her daughters from her posts and those of relatives.
It even suggested prompts asking about their ages and where they live as a family unit.
Privacy concerns
Allegations of displaying deleted photos
Robins also alleged that Meta AI displayed photos of her daughters, including one she had deleted years ago.
Meta has integrated its AI assistant into Facebook, Instagram, WhatsApp, and Messenger for answering questions, writing posts, and generating images.
However, this isn't the first time Meta has faced backlash over privacy concerns.
In July it removed a feature allowing users to create AI deepfakes of other Instagram users after facing severe criticism.
Company statement
'We created this feature to help people'
Responding to the backlash, El-Kassaby said they created this feature "to help people get more information about topics or posts that interest them."
She clarified that when someone clicks on a prompt and uses this feature, the response it generates is based on information the user already has access to.
In Robins's case, she speculated that Meta AI may have pulled from videos where she named her children.