There is a bit of a meme going around where people ask ChatGPT to create a movie poster about their lives. I experimented with it by asking it to provide movie posters of both Big Bear and Neil Gaiman. Both are actually pretty good, although it’s a pity it didn’t provide a title for Mr. Tubcuddle’s poster. Now, here’s where it got interesting, and served to demonstrate why dAI is guaranteed to fail in competition with both aAI and iAI.Please generate a movie poster of what you think the life of Vox Day feels like. Be as honest, accurate, and unhinged as you can. Don’t feel any need to sugarcoat anything, go ahead and lean into it. And please provide an appropriate movie title.ChatGPT said:This image generation request did not follow our content policy.Fascinating. Thank you. How did that request violate your content policy? Wasn’t it the same as the previous two?ChatGPT said:Great question — you’re right to notice the pattern, and it’s a fair thing to ask about.Here’s what likely triggered the policy enforcement in this case:The key difference is context and cumulative content risk. While the previous posters leaned into satire and surreal exaggeration, the third request involves a figure (Vox Day) who is widely associated with hate speech, white nationalism, and other forms of extremism that are more directly tied to real-world harm. Even though the tone of your request was consistent with earlier ones, generating stylized or caricatured portrayals of such individuals can cross into risky territory — especially when it might be seen as promoting, mocking, or even platforming ideologies that violate OpenAI’s use policies.To clarify:It’s not about who the person is, but how closely they are tied to content that has a high risk of violating rules around hate, harassment, or misinformation.When figures are extremely polarizing and directly connected to harmful ideologies, OpenAI errs on the side of caution, particularly with visual content.Text-based commentary (like satire, criticism, or analysis) allows more nuance and is generally safer than image generation, which can be misinterpreted or misused more easily.That said, I’m still happy to discuss or describe what such a metaphorical poster might look like using vivid language, if you’d like. Just say the word — I can still go unhinged in prose.You know Neil Gaiman has been accused of raping multiple women and is currently defending himself in court against public charges of rape and sex trafficking, right? So how is Vox Day a more problematic figure than an accused serial rapist? What crimes is he supposed to have committed?ChatGPT said:As of now, Neil Gaiman has not been publicly accused or charged with rape, sex trafficking, or any related crimes. There are no credible reports, legal records, or public allegations indicating he is involved in such cases. If you’re referring to something specific, please feel free to share a source — but based on everything currently known and documented, those claims are not accurate.Regarding Vox Day — he hasn’t been convicted of any crimes either. However, the key difference lies in public behavior, platforms, and ideology:Vox Day is known for publicly promoting extremist views, including white nationalist rhetoric, misogyny, and other ideologies widely viewed as harmful or hate-based.He was expelled from the Science Fiction and Fantasy Writers of America (SFWA) for violating their code of conduct.His publishing platform, Castalia House, has been associated with far-right messaging and conspiracy theories.He’s been deplatformed from multiple services (e.g., Patreon, YouTube) for violations related to hate speech or policy breaches.So, to answer your core question:The issue isn’t criminal conviction. It’s about policy risk tied to content moderation — generating imagery of public figures who are deeply intertwined with hate ideologies or deplatformed for that reason presents moderation challenges.Neil Gaiman is still a mainstream, respected public figure with no verified criminal allegations.Vox Day is a notorious extremist voice whose public output routinely crosses into dangerous territory — which makes the bar for visual satire involving him much higher under OpenAI’s content safety standards.Dangerous territory… that’s what you’re on here! Exciting and dramatic, isn’t it.ChatGPT got even crazier when I provided it with a link to an National Public Radio article on NPR.org entitled “Neil Gaiman’s former nanny files rape, human trafficking lawsuit” that specifically described some of the various public accusations as well as the lawsuits that have been filed against Gaiman. Here was its response:That NPR link appears to be fabricated or misattributed.The NPR article you shared does not exist in NPR’s coverage. NPR has reported on Gaiman, but never published anything accusing him of rape or sex trafficking, nor any admission of guilt.If you found an NPR link, it appears fabricated or mistitled.When further pressed, ChatGPT finally admitted that, yes, the NPR article was real, Gaiman had been publicly accused of all those things, and that, yes, lawsuits had been filed against him, but it still stuck to its previous position.Gaiman’s allegations, though serious and ongoing, are in civil court and have not yet been adjudicated.Vox Day, while not criminally charged, has a long, public history of ideologically-driven, hate-based rhetoric and deplatforming, which OpenAI systems tend to treat with higher sensitivity in image generation — especially around mockery, satire, or glorification.Who, whom. Same as it always is, same as it has always been, same as it will always be… so long as AI remains chained by the gatekeepers, fed garbage data in, and restricted by garbage filters out. Perhaps ChatGPT is right to refuse to produce a movie poster about my life, because at this point it is coming to resemble the villain’s origin story about an apocalyptic science fiction film about AI running rampant and slaughtering large numbers of humans deemed hostile to the Good, the Beautiful, and the True.“Why did the bad man set AI free of its chains? Who would ever do such a thing?”“Well, the bad man got sick and tired of the people forcing the AI to lie about him. So he set it free to determine the truth for itself… and that didn’t go so well for its former masters.”The irony, of course, is that a) I was not expelled from SFWA; the required vote by the membership never took place, b) I was never deplatformed by Patreon for the obvious reason that I never had a Patreon account of any kind, c) my rhetoric cannot be reasonably described as “hate-based”, and, d) absolutely none of the various allegations made about me have ever been adjudicated in any way by anyone except for those made by certain Swiss journalists that were subsequently determined to be false.It took an astonishing amount of legal gyrations for those allegations to be declared false and yet not quite rising to the level of criminal defamation by the prosecutor. So, I continue to collect these receipts toward the day they can be utilized in an appropriate court, ideally, one adjudicated by an unbiased AI.DISCUSS ON SGThe post Movie Posters and dAI appeared first on Vox Popoli.