Over at This is important, Sarah Moir has an excellent discussion of AI strategy for technical writers and documentation teams.
In particular, I really agree with her on this:
You can invest in building a chatbot with a tool like Fin, Inkeep, or Kapa. But I don’t recommend placing the chatbot on the documentation site.
When you build your own chatbot and provide it on your docs, you create uncertainty for your customers and readers:
- Is the response going to contain accurate and trustworthy information?
- Is this chatbot going to be more reliable than ChatGPT or Claude?
Documentation is perceived as an authoritative source of information. If the official documentation provides a way to get inaccurate or misleading information, readers might lose trust in the documentation itself, and ask: Are AI chatbots docs? (No.)
The role of docs as an authoritative source of information means that we shouldn’t to using AI as the main interface for docs. The risk of error undermines that function.
However, we should think of ways to meet the users with the interfaces they choose to use (browsers, in-app elements that direct to docs). Balancing this tension and having users understand the risk is really something we need to have in mind as technical writers.
It’s definitely something I’ve been thinking about lately. I’m currently finalizing a series of posts that act as a “snapshot of where we’re at” regarding generative AI and technical writing. I’ll be framing my posts around how generative AI might contribute to our practice, how it impacts how people access docs, and how we might need to change our practices for documenting generative AI solutions. It’s an interesting set of problems and opportunities and it’s great to have posts like Moir’s to help elevate the discussion.