because it is such an industry leading company, I think we could end up with some very bad, fair use precedents or laws because Scarlett Johansson is so sympathetic as a plaintiff compared to open AI. And it's, you know, unless they show us some discovery that proves that they really did hire the voice actor and all the rest of it. I mean, this could lead to some very bad precedents for the industry around fair use.
As of November 30, 2025, the Scarlett Johansson vs. OpenAI dispute over the ChatGPT “Sky” voice has not been litigated to a substantive court outcome, nor has it produced specific legislation clearly traceable to a Johansson–OpenAI case.
What has happened:
- In May 2024 Johansson publicly accused OpenAI of using a voice “eerily similar” to hers, said she had hired legal counsel, and that her lawyers sent letters to Sam Altman and OpenAI demanding details on how the Sky voice was created. OpenAI then paused/removed the Sky voice and apologized, while denying that it was intended to imitate her. (cnbc.com)
- Multiple later reports explicitly note that Johansson has not filed a lawsuit against OpenAI; her action has remained at the demand‑letter/PR‑pressure stage rather than becoming a docketed case (e.g., no “Johansson v. OpenAI” in reported litigation). (techtimes.com)
- AI‑related fair‑use and likeness law has evolved since 2024, but via other cases and policy moves (e.g., Bartz v. Anthropic and Kadrey v. Meta on fair use for AI training; proposed or enacted likeness/AI‑disclosure legislation in the U.S. and U.K.), none of which arise from a litigated Johansson–OpenAI lawsuit. Some of these decisions are actually relatively favorable to AI firms on fair use for training data. (apnews.com)
The prediction, as normalized, was conditional: if the Johansson–OpenAI dispute were litigated to a substantive outcome, it would produce more restrictive, unfavorable fair‑use standards for AI/content companies. Because that antecedent (“is litigated to a substantive legal outcome”) has not occurred yet, we have no way to observe whether the predicted consequence about bad precedents would have followed.
It is also still possible (though increasingly less likely with time) that Johansson could file suit in the future, which might then influence precedent or legislation. Until such a case actually exists and is resolved, the prediction cannot be definitively scored as right or wrong.
Given that the key triggering event has not happened, the correct classification is inconclusive rather than right or wrong.