AI image generators pose privacy and consent risks from photo uploads
Data protection experts reportedly warn that AI image generators used to stylize photos store user data and uploaded images, creating risks that photos could be repurposed for deepfakes, non-consensual intimate imagery, or biometric facial recognition without user consent.
What happened
Users of AI image generators upload photos to create stylized versions such as Barbie dolls, action figures, or anime characters. Many generators store these photos along with personal data including IP addresses and email addresses, according to the advocacy group Mimikama.
Data protection experts warn that stored photos could be repurposed for deepfakes, non-consensual intimate imagery, or other misuse, particularly concerning for images of children. Uploaded photos may also appear in responses to other users' queries or be used to train AI systems and biometric facial recognition technologies.
Reported impact
A user's face could appear in unrelated contexts on the internet, including pornographic content or non-consensual intimate images. Stored data including names, locations, and hobbies could be leveraged for phishing attacks or other criminal purposes.
Hamburg's Data Protection Officer warned that biometric facial recognition systems capable of identifying individuals from large crowds could have 'significant consequences,' though such systems have been recently prohibited in the EU.
Index analysis
Users typically do not realize the extent of data they disclose when uploading photos, including personal information beyond the image itself, according to data protection advocates. The ability of AI systems to recognize facial features, age, posture, and emotional state creates risks that photos could be extracted and misused.
Governance controls
Verbraucherzentrale Bayern notes that the scope of affected data depends on what information users provide as input. The EU has recently prohibited biometric facial recognition systems operating on large-scale population surveillance.
Classification
Sources and evidence
Sources not yet published
This record was researched and written by the Index from original reporting. Its sources are still being verified by an editor, and they are listed here once that check is complete. Until then the record carries no reviewer.