ChatGPT developer OpenAI recently released draft documentation exploring the possibility of allowing users to generate porn and other explicit content through its AI technology.
Wired reports that OpenAI has stirred up conversations in the technology industry with recent disclosures in its Model Spec document, laying out how its wants AI products to function moving forward. One notable revelation is the company’s consideration of enabling sexually explicit content generation, with the aim to make this content permissible in age-appropriate contexts. This development comes at a time when AI-generated pornography, or “deepfake porn,” has become one of the most prominent and feared applications of this technology.
Currently, OpenAI’s usage policies strictly prohibit sexually explicit or even suggestive materials. However, a commentary note in the Model Spec sheds light on the company’s explorations of allowing such content. The note emphasizes that OpenAI is “exploring whether we can responsibly provide the ability to generate NSFW content in age-appropriate contexts through the API and ChatGPT,” with NSFW referring to “not safe for work” content.
Breitbart News previously reported that OpenAI’s GPT store was flooded with AI Girlfriend bots, despite those being against OpenAI’s rules.
The Model Spec document outlines different content types that may be deemed “NSFW,” such as erotica, extreme gore, slurs, and profanity. It is unclear how broadly OpenAI’s explorations of explicit content generation encompass – will it be limited to erotic texts, or extend to descriptions or depictions of sexual violence?
OpenAI spokesperson Niko Felix has stated that the company does “not have any intention for our models to generate AI porn.” However, Joanne Jang, who helped write the Model Spec, admitted that users would ultimately make up their own minds if OpenAI’s technology produced sexual content. Grace McGuire, another OpenAI spokesperson, said that the Model Spec aimed to provide more transparency about the development process and gather public feedback on various aspects, including explicit content generation.
The potential loosening of OpenAI’s restrictions on explicit content raises concerns about its moderation capabilities. Last year, Microsoft had to make changes to one of its generative AI tools after a report by 404 Media found that it had been used to create explicit images of Taylor Swift, which were circulated on the social platform X/Twtter.
AI-generated pornography, or “deepfakes,” have led to harassment and nonconsensual imagery, often targeting women and girls. These explicit images have caused significant harm to the targeted individuals, leading to professional and personal setbacks. As OpenAI faces the challenge of porn and explicit content generation, the question is whether the company can effectively prevent bad actors from using their tools to create such material without consent.
Read more at Wired here.
Lucas Nolan is a reporter for Breitbart News covering issues of free speech and online censorship.