Archive — US — techcrunch
Clarifai deletes 3 million photos that OkCupid provided to train facialrecognition AI, report says
Clarifai deleted 3 million photos from OkCupid used to train its facial recognition AI after FTC investigations revealed privacy policy violations. The company acknowledged acquiring user-uploaded images in 2014 without proper consent, leading to legal consequences for both OkCupid and Match Group, despite no fines being imposed on the first offense.
Key points
- Clarifai deleted 3 million photos from OkCupid to comply with FTC investigations.
- The photos were used to train a facial recognition AI model.
- OkCupid provided user-uploaded images in 2014 after requests from Clarifai.
- This data sharing violated OkCupid's privacy policies.
- The FTC did not investigate until 2019 after media reports.
- A lawsuit settlement occurred last month; no admission of wrongdoing by OkCupid or Match Group.
- Clarifai's actions implied that it accessed the restricted data.
Why it matters
- The incident reflects ongoing concerns about data privacy and consent in AI training.
- As of 2025, AI regulations are expected to tighten, emphasizing the ethical use of data.
- This case highlights the delicate balance between innovation in AI and user privacy rights, a trend being addressed globally.
- The FTC's stance may influence future legislation and compliance requirements for tech companies.
The deletion of these photos by Clarifai underscores the importance of transparency and ethical standards in data use, particularly as AI technologies evolve, marking a significant step in addressing privacy violations in the tech industry.