OpenAI has today announced a new policy regarding its artificial intelligence tool, Sora 2. This follows concerns raised by Hollywood production companies and actors' unions over unauthorized generation of celebrity likenesses.
The new generative video model from OpenAI enables users to create videos by entering prompt words. With minimal effort, users can make living or deceased celebrities perform various actions on command, or even have multiple celebrities appear together in the same video. Such capabilities have undoubtedly sparked outrage—Zelda Williams, daughter of the late actor Robin Williams, condemned videos featuring her father's likeness as "stupid things," "disgusting," and "trash on TikTok," stating, "That’s absolutely not what he would have wanted."
Last week, OpenAI was forced to halt the generation of videos featuring Martin Luther King Jr., calling such portrayals of the late civil rights icon "disrespectful." At the time, the company said it was working on protective measures to prevent misuse, but this did not stop users from generating videos involving well-known figures.
Bryan Cranston, best known for his role as Walter White in the TV series Breaking Bad, also expressed concern after his AI-generated image appeared in various forms online—including appearing alongside the late singer Michael Jackson and the copyrighted character Ronald McDonald.
Cranston brought the issue to the attention of the actors' union SAG-AFTRA (Screen Actors Guild‐American Federation of Television and Radio Artists). Union President Sean Austin stated that actors are now facing a problem of "massive misappropriation of personal likenesses." United Talent Agency called for stricter controls or clear compensation mechanisms, pointing out that "using someone's image without consent, attribution, or compensation is exploitation, not innovation." Creative Artists Agency echoed similar criticisms.
This strong opposition was enough to prompt OpenAI to make changes. The company reaffirmed its commitment to strengthening safeguards against such abuse. In a statement, OpenAI said, "All artists, performers, and individuals have the right to decide whether and how their likeness is simulated." Sam Altman, the company's CEO, personally stated that he is "firmly committed to protecting performers' voices and likenesses from misuse."
Cranston expressed satisfaction with the developments and released a statement saying, "I appreciate OpenAI for implementing this policy and improving its safeguards. I hope that OpenAI and all companies involved in developing such technologies will respect our rights to control the use and replication of our voices and likenesses, both personally and professionally."
