An AI wrote this: How Can AI Be Used in UX Research?

As a UX Researcher, it is our job to understand the user's needs and expectations for a product or service. This is a critical factor in creating a positive user experience. But understanding the user is not always easy and can require a lot of time and resources.

Please note that: Most of the text is AI generated by ChatGPT-3 and has only been checked and supplemented to reinforce the message of the human author's position where input is made clear. All illustrations are AI-generated by DALL-E 2. Also, the original article was made in Swedish and then translated to English using Google Translate (to read it in Swedish, switch to the Swedish RI.SE site in the top right corner).

AI has great advantages in UX Research

Prompt 1: Write a blog post using my language on "How AI can be used to enhance the work of a UX Researcher" similar to this format [to which I inserted a previous blog I wrote "by hand"].

AI can be used to automate and facilitate certain aspects of the UX Research process. One of the great benefits of using AI in UX Research is that it can help analyze large amounts of data and identify patterns and trends that would otherwise be difficult to spot.

Here are five areas where AI can play an important role in facilitating the work of a UX Researcher:

Prompt 2: Rewrite the bulleted list a little shorter, but keep the bulleted format.

Automatic data analysis: Use AI to analyze large amounts of data and find trends and patterns.
Natural Language Processing (NLP): Use NLP to automate the identification and categorization of user feedback and reviews.
Image and video awareness: Use AI to analyze user interactions with products and identify problem areas and opportunities for improvement.
User feedback generation: Use AI to generate feedback based on past interactions with products or services.
Chatbots: Use AI to create a chatbot that can help users find information and answer questions. Reduce the burden on customer service and provide users with quick solutions.

Using AI in the UX Research process can, with these methods, help streamline work and make it easier for the UX Researcher to find important insights.

What are the risks of using AI too much as a UX Researcher?

Prompt 3: What are the risks of using AI too much as a UX researcher? And how to avoid it?

Using AI too much as a UX researcher can lead to a number of risks, including:

Loss of human insight: AI can help collect and analyze large amounts of data, but it cannot replace human insight and experience. Relying too much on AI can miss out on valuable insight and understanding that can only come from human interaction and observation.
Bias and discrimination: AI may be prone to reproduce and amplify existing biases and discriminations in data, which can lead to distorted or unfair analysis and recommendations. This, in turn, can affect the UX design and degrade the user experience.
Misinterpretation of data: AI can be limited by data sources and cannot always understand the context behind user data. This can lead to misinterpretation of data and incorrect conclusions that can harm the user experience.

Human touch and user studies will always be needed

To avoid these risks, one should use AI as a complement to human insight and experience. It is important to understand what AI can and cannot do and to use it appropriately. It is also important to be aware of bias and discrimination in data and to take steps to manage and reduce these risks.

Finally, it's important to always validate AI analysis and recommendations with human user studies to ensure the user experience is as positive as possible.

Emil Stenberg

Author's words:

Hearing about the AI DALL-E (the Image Generator) and ChatGPT (the Text Generator) after the summer of 2022, I've been randomly playing with the tools, which has been entertaining but not really productive. But lately I've been trying to find uses for AI in my work as a UX Researcher. To be honest, I think it's smart to know how to best use it professionally. So I asked ChatGPT to educate me a bit and it also resulted in this blog post.

Through the three prompts, I got ChatGPT to produce a text similar to how I would write a typical blog post about how a UX Researcher can use AI in their work. I had to shorten the blog text a bit by asking it to summarize the listed points and raise some risks to nuance and come to the conclusion I wanted.

I believe that AI, specifically generative AI, has an unavoidable impact on how we work and carry out our tasks. But I also think that there will be a need for human intervention for review and reflection in the future.

I had an AI write this blog post for two reasons:

To showcase (mostly as an experiment for myself) that a text-based AI can quickly produce text about a specific area instead of me having to go in and formulate myself adapted to my language use. (Whether the content is accurate, hacked, or just inspired by the input dataset is a whole other discussion that requires a separate debate).
That human behaviors are not predictable and that even AI itself advises that user studies still have a place in the future.