![](https://i.gzn.jp/img/2024/05/10/openai-considers-ai-generated-pornography/00.jpg)
May 10, 2024 13:17:00
OpenAI, the developer of chat AI ‘ChatGPT’ and image generation AI ‘DALL-E’, has revealed that it is considering allowing users to generate content ‘including pornography, gore, and slander’ through its AI products.
OpenAI considers allowing users to create AI-generated pornography | Artificial intelligence (AI) | The Guardian
Until now, OpenAI has prohibited the generation of ‘sexually explicit or suggestive content’ except for scientific or educational materials. However, Mura Mirati, Chief Technology Officer of OpenAI, said at the announcement of the video generation AI ‘
Sora ‘, ‘It is not difficult to imagine that artists are looking for a creative environment where they can express themselves further using AI. We are working with artists and creators in various fields to understand what is useful and how much flexibility the tools should provide.’
OpenAI also revealed that it is considering enabling developers and users to responsibly generate NSFW content through its products. According to OpenAI, the content it is considering may include ‘pornography, extreme gore, slander, and one-sided profanity.’
OpenAI said, ‘We look forward to better understanding user and societal expectations for model behavior in this field.’ In addition, OpenAI’s Joan Zhang reported, ‘OpenAI would like to start a discussion about whether the generation of erotic text and nude images should continue to be prohibited.’ On the other hand, he pointed out that ‘the generation of deep fakes should not be allowed.’
‘We want to give people as much control over their AI as possible without violating the law or the rights of others, but enabling deep fakes is out of the question,’ Jiang said.
In addition, Jean said, ‘Whether the generated text or image is considered pornographic is up to the user’s definition,’ and ‘The discussion about NSFW permission is the conversation OpenAI wants to have.’
‘I am sceptical of any way in which OpenAI seeks to limit the generation of NSFW content to consensual, lawful material,’ said Claire McGlynn, a law professor at Durham University.
In addition, the charity Internet Watch Foundation (IWF), which protects children from online sexual abuse, criticized OpenAI, saying, ‘Despite the recent proliferation of problems such as AI-generated deep fakes, it is disappointing that OpenAI has not taken practical measures and has not taken corporate responsibility for the harm that AI creates. ‘IWF further pointed out that ‘allowing the generation of NSFW content undermines OpenAI’s own goal of developing safe and beneficial general-purpose artificial intelligence.’