Does AI Actually Produce Better Knowledge Base Articles?

AI is so hot right now.

I can't think of another time in tech that something has been so pervasive. Ever since OpenAI launched its API in 2020, AI has been everywhere.

AI has not only permeated task managers and customer support tools, but it has also sparked the emergence of a new wave of startups, all leveraging its potential.

It's not like none of us saw this coming, but I think it did seem like it would be in the distant future rather than so soon. Now it's everywhere in the space of a few years.

Unfortunately much of it seems to have been rushed out the gate, and activists are worried about the ethical and safety concerns that come with it.

The EU launched a strategy in April 2021 to address some of these concerns. The strategy includes encouraging AI development but in a safer, more privacy-focused way.

We ourselves have held back from rushing out an AI. We wrote a post about GPT-3 a while back about its uses but didn't feel like it was quite ready yet.

The world of AI has moved on significantly since then. The models are better, trust has become a larger concern, and we know more about what it can and can't do (and how to write better prompts).

We're at the stage where we're more comfortable with this technology and how to use it. I'm excited to say we're working on something in the background that will make writing articles a whole lot quicker.

But I digress.

In this blog post, I'm going to be exploring whether or not AI is worth using in your Knowledge Base article creation process. Is it a time-saver or a time-sucker?

The Role of AI in Help Documentation

AI's impact on help documentation has been a real game-changer.

At its core, we see AI as a tool to boost customer service teams, not take over their jobs. The big challenge right now is how much time these tasks take up.

When you're creating a new article, you often start with a blank page.

You know what you want to write about, but figuring out how to say it effectively can be tricky.

Decoding the Role of a Modern Technical Writer
Technical writing is more than writing manuals and help documentation. But things have changed with new-age teams and AI.

Using templates (which we call Clips) can speed up the initial draft, but you still have to edit and refine the content yourself.

This is where AI really shines, providing inspiration and generating first drafts.

AI can also dig up valuable insights from large datasets, cut down on complicated jargon, and ensure writers stick to specific style guides and standards.

This not only boosts the quality of the documentation but also makes it easier for readers to understand.

One of the most obvious applications of AI is its ability to create articles.

AI-driven tools can whip up content quickly, letting writers focus more on fine-tuning instead of the grunt work of drafting.

This can significantly cut down the time it takes to produce high-quality documentation, benefiting both the creators and the users.

So, while AI is shaking up the field, it’s key to remember its real value lies in boosting human capabilities, not replacing them.

The Wins and Woes of AI for Your Knowledge Base Article Creation

AI certainly can't replace your brain (just yet, anyway) since you have ingrained knowledge about your company, what your team expects, and what you're trying to achieve.

How to Meet Your Customer Service SLAs With a Knowledge Base
Customer support service-level agreements can benefit everyone, but it can be tricky to keep to the targets. Here’s how to use your Knowledge Base to meet your metrics.

Let's say you need to write an article about using a new feature called Kitty Coins (yep, that came out of my head 😸). Kitty Coins allows users to buy accessories for their virtual cat.

You know that you need to explain:

  • What Kitty Coins are
  • Why you'd want to buy Kitty Coins
  • How to buy said Kitty Coins
  • The terms and conditions of buying Kitty Coins (can they be exchanged/refunded?)

One of the problems with starting off with a blank page is that you don't know quite how to get started. Editing is often a lot easier than writing from scratch.

It's far easier to ask an AI to create the base article for you. You know the outline, you just need some inspiration.

Our content in monthly bitesized emails

Get our best content delivered straight to your inbox.


The trouble is, AI doesn't know what Kitty Coins are. So you have to make sure to explain it. And you have to make sure there's enough detail so it doesn't come up with absolute nonsense.

So you have to be quite careful about what you feed it. You can't say "Explain what Kitty Coins are" because the model will likely make that part up. You'll instead need to write something along the lines of:

You're a technical writer. Introduce what Kitty Coins are (They're virtual currency users can purchase either one-time or as a subscription to spend on virtual cat accessories such as hats, boots, and coats to make their cat look fantastic).

and AI will spit out something like:

Kitty Coins are a type of virtual currency that users can purchase either as a one-time buy or through a subscription service. These coins can be spent on a variety of virtual cat accessories, including fashionable hats, adorable boots, and cozy coats, all aimed at making their virtual pets look absolutely fabulous.


But is it worth the effort to make these prompts when you could've simply written the article in that amount of time? The answer depends.

Staying Inspired With AI Article Creation

One of the tricky parts of using AI in help documentation is the lack of built-in knowledge it has (ironic, right? 😆).

Since the AI doesn't inherently understand your specific platform, how it works, or why customers would want to use it, you basically have to teach it by crafting a better prompt. This is often called Prompt Engineering and it takes a fair bit of skill and nuance.

At this point, you might be wondering what the point is.

Why bother with prompt engineering when you could just write the article yourself from scratch? 🤷

It's a valid question, and one many have asked. I think the answer lies in the potential for inspiration.

As I've been testing out our own AI feature (🤫), I've found writing prompts can be a bit annoying and tedious. Despite this, it’s proven to be useful for generating inspiration.

The Ultimate Guide to Creating Your First Knowledge Base Article
Creating your first Knowledge Base article can feel like a giant hurdle. Have no fear—we’re here to help you get started. Read on to learn how to go from zero to KB hero.

Sure, I've had to go back and edit the articles to remove the nonsensical or irrelevant parts the AI sometimes produces. But amidst these edits, I've often found sparks of inspiration that I wouldn't have encountered otherwise.

In some of the articles I've generated, there are sections I never would have written or even thought of myself. Sometimes this spontaneous generation leads to content that isn't useful, but more often than not, it opens up new avenues of thought.

This has been particularly valuable because it has expanded my horizons and introduced me to ideas and perspectives I might not have considered on my own.

I don’t claim to be a great or even particularly good writer, but working with AI has shown me aspects I could have included in my articles that I would never have thought of before.

I believe this collaborative process between human and machine has ultimately made the Knowledge Base articles better and more comprehensive.

In the end, while AI may not replace the need for human writers, it can definitely complement and enhance our capabilities, fostering creativity and improving the quality of the content we produce.

Replacing the More Mundane Tasks in Your Knowledge Base Article Process

By automating repetitive tasks and offering smart suggestions, AI helps technical writers focus on more complex and meaningful work like editing, updating, and enhancing articles.

Take, for example, writing metadata like SEO descriptions and creating search tags. It’s not the most exciting part of writing an article, often seen as a necessary but tedious step 😅

Crafting the perfect metadata requires creativity and a good understanding of search engine algorithms and user behavior.

This is where AI for your Knowledge Base really shines.

AI tools can quickly and accurately generate meta descriptions and search tags based on your article’s content. These tools analyze the text, identify key themes and phrases, and generate suggestions that are both relevant and optimized for search engines.

Optimizing Your Knowledge Base Search
Discover tips for optimizing Knowledge Base searches. Learn how categorizing a Knowledge Base makes search easier and avoid common issues with proper optimization.

This not only saves time but also ensures your metadata aligns with the latest SEO practices, improving the article’s visibility.

Let's go back to the Kitty Coins example.

If you’re writing about Kitty Coins—the virtual currency used to purchase cat accessories in an online game—AI can suggest meta descriptions like “Learn how to use Kitty Coins to buy virtual cat accessories” and tags such as “virtual currency,” “cat accessories,” and “subscription service.”

These AI-generated suggestions are almost good enough to use and tailored to boost the discoverability of your content.

Using AI for these tasks can significantly boost your article’s visibility and searchability, making it easier for readers to find the information they need.

Automating these mundane tasks allows writers to invest more time in creating detailed, engaging, and high-quality content. Plus, AI can continuously learn and adapt, improving its suggestions over time as it gathers more data on user interactions and preferences.

So, while you focus on crafting rich and engaging content, you can let AI handle the behind-the-scenes optimization to help your articles reach a wider audience efficiently.

Embracing AI in your content creation process not only streamlines your workflow but also elevates the overall quality and reach of your Knowledge Base.

In the end, both technical writers and readers benefit from a more streamlined and effective content experience.

AI in your Knowledge Base might not mean it's creating super useful content right away, but it might mean you write better articles with its assistance.

Use AI as an Assistant, Not a Replacement

Embracing AI in your content creation process can truly reframe the way you approach writing Knowledge Base articles.

While it might sound daunting at first, the benefits are hard to overlook. From enhancing your creativity to automating mundane tasks, AI offers a valuable toolset that aids in producing higher quality, more engaging, and optimized content.

How to Create a Seamless External and Internal Knowledge Base Experience
Creating a great documentation experience for your customers and team is tricky. Here’s our insight into how to make it as simple as possible.

Sure, there’s a learning curve when it comes to prompt engineering and fine-tuning the AI to meet your specific needs. But, once you get the hang of it, the AI becomes an indispensable collaborator that can spark new ideas and ensure your articles are as polished and discoverable as possible.

The most rewarding aspect is the collaboration between human writers and AI.

This partnership not only enhances the quality of the content but also makes the whole process more enjoyable and efficient.

So, go ahead and explore how AI can assist you in creating a more robust, dynamic, and user-friendly knowledge base. You might just find that it’s the perfect partner in your writing journey ✍️✨