Meta’s Oversight Board, an independent policy council, is delving into how Meta’s social platforms handle explicit AI-generated images. The board announced on Tuesday that it’s investigating two cases involving Instagram in India and Facebook in the U.S., concerning Meta’s failure to detect and respond to explicit content.
In both instances, the platforms have since removed the content. To avoid gender-based harassment, Meta chose not to disclose the identities of the individuals targeted by the AI images.
The Oversight Board examines cases related to Meta’s moderation decisions. Users must first appeal to Meta regarding a moderation action before approaching the Oversight Board. The board will publish its comprehensive findings and conclusions in the future.
Case Details:
In the first case, a user reported an AI-generated nude image of an Indian public figure on Instagram as pornography. Despite multiple reports, Meta failed to remove the image promptly. It was only after the user appealed to the Oversight Board that Meta took action, citing a breach of its community standards.
The second case involves Facebook, where an explicit AI-generated image resembling a U.S. public figure was posted in a group focusing on AI creations. Although the image was promptly removed, the Oversight Board selected this case to assess Meta’s global policy effectiveness.
Addressing Deepfake Porn and Gender-Based Violence:
Generative AI tools have facilitated the creation of pornographic content, particularly concerning in regions like India. The Indian government has expressed dissatisfaction with tech companies’ handling of deepfakes, emphasizing the need for robust legal frameworks.
Experts emphasize the importance of regulating AI models to prevent the creation of harmful content. While there are few laws addressing AI-generated porn globally, efforts are underway to criminalize such activities.
Meta’s Response and Next Steps:
Meta acknowledges taking down the objectionable content but hasn’t addressed its initial failure to do so. The company utilizes a combination of AI and human review to detect sexually suggestive content, aiming to limit its distribution.
The Oversight Board seeks public comments on the matter, focusing on the harms of deepfake porn and Meta’s approach to detecting AI-generated explicit imagery. These cases underscore the challenges platforms face in combating harmful content while adapting to AI advancements.
