OpenAI has released new policies for its Sora 2 video generation model following extensive criticism from Hollywood studios and actors’ unions. The backlash centered on the tool’s ability to create likenesses of individuals without their consent.
The Sora 2 model allows users to generate video content from text prompts, which has enabled the creation of videos featuring famous figures, both living and deceased. This capability drew public condemnation from Zelda Williams, who criticized AI-generated videos of her late father, Robin Williams. She described the content as “dumb,” “disgusting,” and “TikTok slop,” and stated that creating such videos was “not what he’d want.”
In the week preceding the policy announcement, OpenAI had paused the generation of videos featuring Martin Luther King Jr., labeling them “disrespectful depictions” of the civil rights leader. Despite company statements about working on preventative guardrails, users continued to generate content with well-known figures. Actor Bryan Cranston voiced concern after an AI-generated image of him appeared online alongside the deceased singer Michael Jackson and the copyrighted character Ronald McDonald.
Cranston escalated the issue to the actors’ union SAG-AFTRA, where President Sean Astin stated that actors now confront a “massive misappropriation” of their identities. United Talent Agency called for increased controls or financial compensation for performers, arguing that the “use of such property without consent, credit or compensation is exploitation, not innovation.” Creative Artists Agency also expressed similar criticism regarding the technology’s misuse.
In response to the widespread backlash, OpenAI announced it would strengthen the protective guardrails for its model. The company issued a formal statement affirming its new policy: “All artists, performers, and individuals will have the right to determine how and whether they can be simulated.” OpenAI CEO Sam Altman reinforced this position, declaring he is “deeply committed to protecting performers from the misappropriation of their voice and likeness.”
Following the announcement, Bryan Cranston issued a statement acknowledging the policy adjustments. He stated, “I am grateful to OpenAI for its policy and for improving its guardrails, and hope that they and all of the companies involved in this work respect our personal and professional right to manage replication of our voice and likeness.”
