OpenAI has released preliminary documentation detailing its behavioral expectations for ChatGPT and other AI technologies. The documentation discloses that the company is contemplating the inclusion of pornographic and other explicit content, despite its current policy prohibiting sexually suggestive materials.
The company is considering the responsible provision of NSFW (not safe for work) content within appropriate contexts through its API and ChatGPT. This exploration raises questions about whether OpenAI will slightly adjust its usage policy to allow for the generation of erotic text, or if it will broaden its scope to include violent descriptions or depictions.
In response to inquiries, OpenAI spokesperson Grace McGuire stated that the Model Spec aims to increase transparency and gather feedback from various stakeholders. The company’s Chief Technology Officer, Mira Murati, expressed uncertainty about allowing nudity in future video generation tools.
AI-generated pornography has emerged as a major concern, with deepfake porn being a prevalent form of harassment. OpenAI’s potential foray into explicit AI content has raised alarm among experts, highlighting the risk of intimate privacy violations and nonconsensual use of synthesized images.
While OpenAI’s policies prohibit impersonation without consent, the challenge lies in effectively moderating explicit content generation to prevent misuse. Microsoft previously adjusted its generative AI tools after reports of explicit content creation and distribution.
The implications of OpenAI’s exploration into explicit content generation remain to be seen, with concerns about moderation and potential misuse. The company’s deliberation on this matter underscores the complex ethical challenges associated with AI technologies.
