Meta changes AI prompts after invasive questions about user's children
Meta is changing AI prompts after a viral video showed its chatbot asking invasive questions about a user's daughters. It missed the mark.
The change follows a video posted last week by Instagram user Kalie Robins, who said Meta AI presented her with an invasive suggestion after she cross-posted a clip to Facebook. The prompt 'Who is the child passenger?' appeared beneath a video of her and her child singing in a car, according to The Verge.
When Robins selected the prompt, she said Meta AI pieced together information about her daughters from her previous posts and posts from her relatives. It then suggested additional prompts asking about her children's ages and where Robins lives with her family. Robins said Meta AI also showed photos of her daughters posted to the platform, including one she claims she deleted years ago.
In a statement to The Verge, Meta spokesperson Dina El-Kassaby said the company "missed the mark" and that "the feature never should have prompted the individual with questions like that." El-Kassaby said Meta has fixed the issue that caused Meta AI to suggest prompts related to personal topics.
El-Kassaby said the feature was created to help people get more information about topics or posts that interest them. When someone clicks on a prompt and uses the feature, she said, the response it generates is based on information the user already has access to. In this case, Meta AI may have pulled from videos where Robins named her children, according to El-Kassaby.
Meta has built its AI assistant into Facebook, Instagram, WhatsApp and Messenger, where it can answer questions, write posts and generate images. In July, the company pulled a feature that allowed people to create AI deepfakes of other Instagram users after facing fierce backlash.