Tag: SocialMediaSafety

After investigation, AI-generated sexualized avatars are removed from social platforms Problem Overview Multiple accounts have been removed from social media sites after complaints were made about AI-created female avatars being used in a harmful or misleading way. The accounts featured digital characters with a high sexualization factor that looked like Black women. Some profiles use misleading identities and suggestive images to quickly attract large audiences. Some profiles also advertised links that led users to external websites with paid adult content. The Accounts Worked These accounts used AI-generated images and videos to make them look like actual influencers. The avatars are often exaggerated and have unrealistic skin tones. Some accounts names contained words related to race or appearance. Some posts had captions with provocative and racialized topics. These accounts had been linked through likes and follows, creating a social network which increased visibility. Many profiles claimed that they did not use AI even though their content was completely generated. The Impact of Real Content Creators In one case, a video of a real-life model was altered by AI in order to substitute her face with an artificial avatar. The original footage and movements of the model were used without her permission. The content of her profile was viewed far more on the fake profiles than it did on her real account. These videos were viewed by tens and millions. The misuse was reported to her multiple times, but it took a long time before any action was taken. She was concerned about identity theft as well as the spreading of false content. Platform Response Several accounts were deleted after the problem was brought to light. However, similar profiles had already spread across multiple platforms. Several companies said they are reviewing AI-generated content and enhancing detection systems. Expert Concerns Digital analysts and researchers warn that AI-generated avatars may reinforce stereotypes. The content is often based on exaggerated features of race and sexualized frames. AI also makes it simpler to distribute and create synthetic identities in large numbers, according to experts. Virtual personas are not accountable to real-life social consequences. Also, users may find it increasingly difficult to distinguish between AI-generated content and real material.
Crime and Justice

After investigation, AI-generated sexualized avatars are removed from social platforms

Problem Overview Multiple accounts have been removed from social media sites after…

Register Lost your password?