OpenAI Considers Lifting Ban on AI-Generated Explicit Content

Currently, non-open-source AI applications that generate images based on text prompts prohibit users from creating explicit content. If users attempt to input related keywords, the system may automatically intercept and block the content, and in severe cases, even ban the user's account.

This consensus is widely followed by various AI models, at least for now. However, open-source or locally runnable AI models do not have such restrictions, allowing users to deploy and generate any content they want.

In response to user expectations, OpenAI is exploring whether it's possible to generate NSFW (not safe for work) content in a responsible manner, within age-appropriate environments.

It's worth noting that OpenAI's consideration goes beyond just image generation. The company is also discussing the possibility of using AI models like ChatGPT to generate explicit text content, which would require age verification to ensure users are of legal age.

For image generation, while OpenAI is considering allowing users to create explicit content using AI (including potentially generating explicit videos using the OpenAI Sora model in the future), the caveat is that users cannot create deepfake content that impersonates others.

These discussions are still ongoing, and the company aims to ensure that users have maximum control over their content without violating laws or infringing on others' rights.

OpenAI emphasizes that discussing these topics does not mean the company is actively developing AI-generated explicit content. The decision to lift these restrictions will be made after the discussion is concluded.