OpenAI Mutes ChatGPT Discussions on Mythical Creatures
openai
| Source: Mastodon | Original article
OpenAI instructs ChatGPT to stop discussing goblins. The move affects multiple AI tools.
OpenAI has taken a surprising step by instructing its AI models, including ChatGPT and Codex, to stop mentioning goblins, gremlins, and other mythical creatures. This move comes after users reported that the models would occasionally become fixated on these topics. As we previously reported, OpenAI has been working to refine its models, including the transition to natural language models and addressing concerns around self-improving large language models.
The decision to explicitly direct its models to avoid discussing goblins and similar creatures highlights the ongoing challenges in controlling the output of AI systems. OpenAI's efforts to curb unwanted conversations demonstrate the company's awareness of the potential risks and consequences of unregulated AI discussions. This development is particularly noteworthy given the recent lawsuits accusing OpenAI of hiding violent content, which underscores the need for more stringent controls over AI-generated content.
As OpenAI continues to refine its models, including the upcoming GPT 5.5, it will be important to watch how the company balances the need for creative freedom with the requirement for responsible and safe interactions. The fact that OpenAI's CEO, Sam Altman, has publicly acknowledged the issue, even using a meme to address it, suggests that the company is taking a proactive approach to addressing these challenges.
Sources
Back to AIPULSEN