Using Gen AI for Technical Communication: My first steps

You are at the beginning of your Technical Writing career path. You haven’t yet tried using Gen AI tools. Or you are curious to learn more about the experiences of fellow writers. You are in the right place!

In this article, I’m sharing the challenges I had related to the use of Generative AI tools (further referred to as Gen AI assistants). I’m describing the ways I used to overcome those challenges to advance in my current role of a Technical Communicator.

By the way, did you know that according to 2024 Work Trend Index Annual Report from Microsoft and LinkedIn: “AI is being woven into the workplace at an unexpected scale. 75% of knowledge workers use AI at work today, and 46% of users started using it less than six months ago”?

When I heard about ChatGPT in 2023, my first thought was: “Oh my! What will we—people in general and Tech Writers in particular—do? This technology will take over our work, and we can’t help it anyhow!”

To get over my anxiety, I decided to dive deeper into the world of ChatGPT, which was quite challenging for me in the first place. You might call me conservative or old-school, but I need to understand the basics before trying new technologies. Additionally, due to security precautions, I had absolutely no idea of how to use the tool: what questions to ask and how detailed my questions to ChatGPT should be.

The first call

Luckily for me, our company had launched the initiative where colleagues from different departments had an opportunity to test the capabilities of the new technology. “I can do something helpful!”, I thought.

To begin with, I carefully listened to my colleagues who had started using ChatGPT. They were so excited about it. So, I started digging. I read some articles, listened to some podcasts, and watched some video tutorials about it. And finally, I was ready to try it by myself.

Having access to the paid version of ChatGPT, I had no need to think about the security measures. However, wait! I did think about it. I was very cautious about sensitive data. I’ve never added and still never feed ChatGPT with personal names, product names, or any company- or customer-specific data.

My first approach to ChatGPT for Technical Writing was cautious. I didn’t know what to ask or how to properly formulate my questions. I always wondered what I wanted to get as a response. The reason for that, I think, was that I had no real need to use it. I was playing around asking some generic questions and received some generic answers. For example, “What is Technical Writing? What do Technical Writers do? How to write good technical documentation? Etc.” And that was it.

My conclusion about using ChatGPT was that there was nothing special about it. I was wondering why there was much fuss around it. Why were so many people extremely excited about it? Why were so many companies putting so much effort into developing AI-based solutions? But the story didn’t end there.

The real need for a Gen AI assistant

On my project, I had to do tasks that required more domain knowledge than I had. I googled a lot and had to go through numerous articles, most of which were not helpful. And one day in January 2024, I received access to another Gen AI tool called Microsoft Copilot. “Why not try it?”, I thought.

I had 15 requests per day on Copilot, which was more than enough for me then. I asked questions about the area I worked with—definitions of terms, existing solutions, latest research outcomes and trends, experts’ predictions for the future, and others.

Copilot suggested answers with the links to the corresponding articles. And that was a wow! It came to me that Copilot really saved my precious time. I could read a short answer to the question or read the whole article if it was relevant. And if not, I could just skip the link and save myself from diving into the deep hole of unrelated knowledge.

Playing with Copilot helped me understand the core difference between surfing the web and asking Gen AI. And it’s so easy: Google suggests links leading to tons of materials, while Gen AI assistants suggest straightforward answers to your questions. Since then, I started using Copilot from time to time.

Gen AI for Technical Communication

The first thing I want to highlight is that regardless of your knowledge of English language or the guidelines and rules of Technical Writing, it’s nearly impossible to do your job without an understanding of the area you are writing about—your domain knowledge (e.g., FinTech, Big Data, Internet of Things (IoT), Machine Learning, etc.). And here I mean lots of things like:

• The knowledge you get from studying (i.e., school, university, courses, certification programs).
• The knowledge you get from subject matter experts (SMEs).
• The knowledge you get from the Internet (social media posts, articles, books, and other).

That is why I used Gen AI for enhancing my domain knowledge first.

At some point I realized that asking assistants questions was not enough. As any journey has at least a rough plan, professional communication with Gen AI assistants needs it as well. And here prompt engineering steps in. New research. New findings.

That experience was fantastic! I had very clear goals:

1. Research the available educational materials related to Gen AI and prompting.
2. Get hands-on experience by experimenting with prompt requests.
3. Come up with the best prompts for creating a PowerPoint presentation structure.

That’s where the world of Gen AI “shined bright like a diamond”. I asked three different tools—ChatGPT, Copilot, and our corporate AI-powered chatbot—lots of questions, including their advice on proper prompting.

The answers differed from the first glance. However, all the assistants provided similar concepts and ideas. From my perspective, they added to each other. I analyzed the similarities and came up with a prompt template for a presentation structure. For now, it looks ideal to me to begin with.

I’m using this prompt template when working on the structure of a new presentation. And here it is:

Act as a seasoned Technical Communicator. Create a structure of the presentation about <your_topic>. The target audience is <your_audience> who are just in the beginning of their career and know nothing about <your_topic>. The presentation should have <number> slides.

It would be hard to share all the communication I had with different Gen AI assistants in this article. But believe me, there was so much of it— communication.

Using prompts saves me from reinventing the wheel each time I need to do a specific type of task. However, let’s remember that each task is unique. Thus, it requires more specific requests as a follow-up to the first response you get from an assistant.

All these experiments with ChatGPT, Copilot, and others helped me understand that this new and fast-developing technology is worth all the excitement around it.

Conclusion

Summarizing, I can say that Gen AI assistants in Technical Communication help to:

• Save time for understanding domain area.
• Prepare for communication with SMEs better.
• Generate ideas for document content, structure, architecture.
• Test the clarity of the content before sending it to the peer or SME review.
• Write scripts for videos.
• Create API documentation.
• And many more.

These are the benefits that first come to my mind. Meanwhile, there are so many cool things Gen AI can do to help us create the best user assistance.

What is more important—Gen AI assistants are a help rather than a threat. They cannot replace people because the only thing they can do is suggest. We are to decide whether to follow the suggestion, search for more ideas that would work better, or come up with something absolutely new as a result of chatting with an AI tool.

In conclusion, I suggest you reflect on several simple things you can do to start using Gen AI tools and improve your skills in Technical Writing, Information Development, Technical Communication:

1. Embrace a fearless attitude towards trying new things. Every experience is important and provides you with new knowledge. This way, you widen your horizons both professionally and personally.
2. Ask questions—it’s a perfect way to learn something new especially if you are chatting with Gen AI assistant. When you read its answers, you understand how clear, specific, and to the point your questions are. This way, you improve your communication skills as well.
3. Try different tools—ask the same questions and compare the results. You will be amazed with the variety of answers different Gen AI assistants can suggest to you. This way, you increase angles of perspective. “So many men, so many minds”, as they say.

I can only imagine how this technology could simplify my routines when I was starting to write technical documentation. And how about you? Have you tried using Generative AI tools for technical documentation writing? What’s your feedback about it?