It's important for general-purpose use: with generative models there's always a chance of hallucinations. For all uses except the specific adult-flavoured ones,

It's important for general-purpose use: with generative models there's always a chance of hallucinations. For all uses except the specific adult-flavoured ones, you don't want the response to contain vulgarities. No one wants their company's chatbot to start narrating furry erotica. If trained on adult content, you would need to have a burdensome moderation layer downstream of the LLM. When you do want the more niche adult themed LLMs, there are fine-tuning datasets available. Fine-tuning a vanilla open-source LLM for these uses works great. There are active communities of adult roleplay LLMs on imageboards.