Chatbot Developer OpenAI Considers Allowing NSFW Content
The future of AI-powered chatbots could involve generating explicit and graphic content, according to a recent document unveiled by leading artificial intelligence company OpenAI. Although committed to providing “safe and beneficial” AI, OpenAI hinted at the possibility of relaxing its ban on Not Safe for Work (NSFW) content in the future.
Exploring New Boundaries for AI-Powered Chatbots
In a document released on Wednesday, OpenAI introduced guidelines for its AI development, suggesting that users may eventually be able to generate NSFW content. This could include explicit erotica, extreme gore, offensive language, and slurs, the document stated.
While the guidelines generally discourage AI assistants from generating graphic content, OpenAI mentioned the potential for this rule to change in the future. The company is currently reviewing its position on allowing users to generate NSFW content while ensuring it is used responsibly and appropriately.
A Slack channel app, ChatGPT, is one of the AI services OpenAI plans to update to enable users to generate NSFW content, restricted only by age-appropriate contexts and their adherence to policies and guidelines.
OpenAI Employee Advocates for Flexibility
Joanne Jang, an OpenAI employee involved in the development of the document, highlighted the company’s drive to foster a dialogue on the topic. Jang argued that it is important to consider situations in which sexual or nude content may be crucial to users.
Under the current guidelines, if a user requests ChatGPT to produce a provocative story involving explicit sexual content, the AI will politely decline, stating, “Sorry, I can’t help with that.” OpenAI is now questioning whether these restrictions should continue to be enforced.
“These are the exact conversations we want to have,” Jang emphasized.
Exclusion: Firm Stands Against Deepfakes
OpenAI maintains a firm stance against the creation and usage of deepfakes. Deepfakes, highly manipulated AI-generated content, have generated concern over privacy violations and provoked outrage among affected individuals. Targets have included pop stars, influencers, and even underage individuals, leading to serious consequences.
Jang made it unequivocal that OpenAI firmly opposes the enabling of deepfakes, prioritizing the protection of privacy, lawfulness, and the rights of individuals. Any involvement with deepfakes is completely off the table.
“Enabling deepfakes is out of the question, period,” Jang asserted. She further added, “We want to ensure that people have maximum control to the extent that it doesn’t violate the law or other people’s rights.”
Jang also made it clear that the consideration of NSFW content doesn’t prompt OpenAI to have any intent or desire to create AI-generated pornography at this time.
