Part of:Using generative AI writing tools in business workflows
Will AI replace technical writers? One user's experience
As with many jobs, how AI advancements will affect the future of technical writing is up for debate. Hear from one expert on why AI is here to stay -- but so are technical writers.
As a courseware author, technical writer and lab developer, I've watched the emergence of generative AI with great interest. Will it replace me, destroying the small business I've spent years building, or will it be a tool I can use to provide even more value to my customers?
I certainly hope for the latter option, and I think it's a realistic one. Experienced writers -- and creatives in general -- really can't be replaced by AI. There's a depth of human experience, intuition and insight that AI simply cannot fully replicate. That said, writers who ignore AI are missing out on one of today's most important tools.
To understand the relationship between technical writers and AI, we must first consider the human elements of technical documentation. Rather than feeling threatened by AI, technical writers can use it to enhance their output, working more efficiently and with greater accuracy.
Technical writers' workflows and AI
A technical writer's primary role is to create clear, concise and accurate documentation on complex content. Technical writers work with subject matter experts to thoroughly understand a technology and tailor explanations to a target audience.
The role requires expertise in planning, editing, multimedia integration and collaboration to generate useful and clear resources. In addition, technical writers often update material based on user feedback, product development and new technologies.
Will AI replace technical writers?
There's been some discussion about whether AI makes the technical writer role obsolete. A clear understanding of the role indicates this is not the case, and we should expect technical writers to remain a critical part of product support.
AI is a tool in the technical writer's toolbox, not a replacement.
For one thing, AI tools won't necessarily have information about what technical writers are documenting -- the technology in question might be new, proprietary or even experimental. In those scenarios, only a person with access to the engineering or development teams can create the necessary reference material.
AI is a tool in the technical writer's toolbox, not a replacement. A human touch is still necessary for providing context, relatable examples and logical instructions.
Roles and titles in IT often change as quickly as the technology ecosystem itself, but the need for human supervision remains. Technical writing is a great example of this.
AI as a tool for technical writers
Instead of replacing writers, AI can make technical writing more effective by making it easier for us to do our jobs -- and do them well.
AI can enhance research, creativity and proofreading in technical writing in several ways:
Research. AI-powered search tools enable more informed writing by improving access to accurate and current information. These tools also increase the efficiency of the research process, helping authors find relevant information more quickly.
Creativity. AI expands creativity by letting authors experiment with different approaches to delivering content through their queries, often from unique or unexplored perspectives. AI can also provide prompts, data and inspiration for new pieces, strengthening an author's ability to write interesting content.
Proofreading. AI tools make the writing and editing process quicker and more effective by checking spelling, evaluating grammar and generating citations.
AI limitations for technical writers to know
While AI can be a useful tool, it also has significant limitations that are particularly concerning for technical writers.
Technical writing projects usually involve very accurate explanations and detailed descriptions of procedures for accomplishing a goal. For example, consider documentation supporting the deployment of a major cloud service. This requires a thorough understanding of the process; input from developers, administrators or other subject matter experts; and careful step-by-step instructions.
One major drawback of generative AI is its limited ability to serve as an accurate source of information. This limitation is especially problematic for emerging technologies that lack a large body of supporting information -- exactly the area where technical writers often create content.
Concerns surrounding AI for authors of technical documents include the following:
AI hallucinations. AI can return false or nonsensical results due to improper training or because of the probabilistic nature of large language models. These so-called AI hallucinations can present incorrect information as facts, misleading a tool's users if the hallucinations make their way into technical documentation.
AI bias. AI can exhibit bias based on its training data and source material, leading to skewed or imbalanced outcomes. These biases might stem from the assumptions of the humans who designed the AI or from nondiverse or biased data sources.
These limitations can lead to situations where AI tools return inaccurate or incomplete data during a technical writer's research phase. Human oversight and critical evaluation of AI source material are essential to mitigate these issues.
Examples of AI tools for technical writers
Many AI tools can enhance a technical writer's process. Generic LLMs like ChatGPT and Claude are many users' first experiences with generative AI, but tools such as Perplexity AI and Consensus AI have specialized features that can be especially useful. Writers who master these tools and integrate them into research and planning can deliver stronger and more efficient documentation.
Perplexity AI
Perplexity AI is a conversational AI search engine that lets users enter queries using natural language. Unlike traditional search engines, Perplexity then returns a summarized overview based on information it finds online.
This summary is often more useful than the tens or hundreds of thousands of links Google returns. It breaks down results into bullet points and other easily consumed formats to explain a concept or technology without sifting through multiple webpages.
Figure 1. Perplexity summarizes and cites information from many sources.
One of Perplexity's most valuable features is its citations, shown in the screenshot above. Rather than just presenting information, it also includes sources, enabling users to verify information or delve deeper into a subject.
This is critical for technical writers using Perplexity as a research tool, as they must be able to follow these citations to confirm the accuracy of results or pursue additional lines of investigation. While it might appear at first glance that a documentation team could simply input a query into Perplexity to generate results, those results would be of questionable value and possibly difficult for readers to consume.
Consensus AI
Consensus AI is a search engine similar to Perplexity, but with a deeper academic focus. In line with its emphasis on academic research, Consensus queries peer-reviewed scientific materials to provide content supported by specific evidence.
Figure 2. Consensus summarizes and cites content from academic material.
This ability to review literature within a specific field makes Consensus a critical tool for technical writers in disciplines like science, medicine and academia. Consensus offers various options for narrowing the scope of searches to pull results only from desired sources.
Closing thoughts
I regularly use Perplexity as a research tool and to develop topics for technical articles, opinion pieces and learning resources. Like Microsoft Word, Grammarly and even computers themselves, AI is just another tool that enhances the technical writer's role. However, writers still control the final product, corralling the information provided by AI tools to maintain accuracy.
Will AI replace technical writers? I don't think so. However, AI tools could become indispensable in creating documentation, training courseware, reference materials and other sources of information.
If you're a writer reading this, consider how you've integrated AI into your process -- perhaps as a tool for research, outlining or proofreading. If you haven't started yet, today's a good day to begin.
If you're an executive aiming to cut costs by reducing head count, be very cautious about letting go of skilled writers who work with subject matter experts, users and product owners to generate the documentation supporting your company's technology. These authors, as a direct connection to your customers, are an essential part of a growing, AI-driven enterprise.
Damon Garn owns Cogspinner Coaction and provides freelance IT writing and editing services. He has written multiple CompTIA study guides, including the Linux+, Cloud Essentials+ and Server+ guides, and contributes extensively to TechTarget Editorial and CompTIA Blogs.
