AI
Meta chief AI officer’s Instagram post of deepfake Sam Altman video draws attention
Meta chief AI officer Alexandr Wang posted an AI-generated video on Instagram showing Mark Zuckerberg, Sam Altman, and the mascot for Meta’s consumer agent Muse in a meeting. The video is labeled as AI content, and Meta said Altman had previously made his likeness available for AI generation, but the incident has renewed debate over protecting public figures’ likenesses.
The short clip was used to promote Meta’s consumer AI agent, Muse. In the footage, Wang, Meta CEO Mark Zuckerberg, OpenAI CEO Sam Altman, and Muse’s anthropomorphic mascot Jolly take part in a four-way video call; Altman joins wearing a fuzzy hooded outfit and face paint similar to Jolly’s. The caption references OpenAI’s product strategy and was seen as a marketing move centered on competition among AI agents.
The Verge noted that just a few years ago, AI content depicting real people often drew public condemnation or even legal threats, whereas today such content has become part of social media reality. The report also noted that political groups ran millions of dollars’ worth of AI-generated attack ads before the U.S. midterm elections.
Meta policy requires AI-generated videos depicting real people to carry an AI label, and the video posted by Wang is labeled; however, static images do not face the same requirement. Previously, AI-generated images of actor Mia Goth circulated on Instagram, and her representatives criticized the platform for not prioritizing protection of public figures’ likenesses, noting the images were also shared by fan accounts with blue verification badges.
Meta spokesperson Tracy Clayton responded that Wang’s post was labeled as AI content and pointed out that Altman had made his image available for AI content generation when OpenAI launched Sora, after which large numbers of fake Altman videos quickly appeared on Sora. OpenAI did not comment on the report.