Using AI in Product Information. Interview with Ivan Bernal, project manager from Sigma Technology Information. Part 2

How is AI changing product information?

There are large language models (LLMs) that can be connected to information about any product. I think in the future, these models will become more powerful and they might read into documentation. I want to be clear – they will not replace technical writers who create manuals. I see their role as interpreters that will explain instructions in a more human language. For example, people will not be limited to specific keywords like “trunk” or “boot” to get necessary information. They will be able to ask, “How do I open that big storage space in the back of my car?” and get the answer. And I think it’s a good development.

Do you know any examples of companies already implementing AI in their manuals?

I can see in the car industry, some companies use AI to give users information. They set a proportion of “hot” and “cold” answers to manage the freedom AI has in their responses. At some point, car companies will figure out the perfect formula for how to give both accurate and user-friendly information with the help of AI.

You said AI can’t replace experts in product information. Why?

AI understands code but it doesn’t understand more complex systems. It cannot look at unconnected parts, analyze them, systematize them, and create a taxonomy of terms. Only an experienced information architect can do that. Neither can AI understand rotating bolts. Technical experts can look at the bolts and understand how they work. They can explain it to technical writers who can comprehensively write how the bolts rotate. Yes, we can use AI-powered tools to help us write smoother texts without mistakes, but we can’t ask AI to write the manual in the first place.

What skills are essential to stay relevant in product information?

The skills that AI is poor at: information analysis, systematization, and classification. Learning about new technology and actual technical knowledge is very useful, too. As I mentioned before, AI doesn't understand how machines work, but you can.

How do you keep these skills sharp?

I analyze information as part of my job every day and learn about technology by attending lectures and researching online. Also, it helps to keep your eyes and ears open for what’s going on at the company where you work.