How Much is AI Used in UX?

Did the student write the paper with AI? Did the teacher review the paper with AI?

Did the respondent use AI to answer the survey? Did the researcher use AI to detect responses that were AI generated?

Was this article written with AI? Was the LinkedIn post about this article written with AI?

There’s hardly a hotter topic in the UX field than the appropriate use of AI in research and design.  The conversation has shifted from whether AI will take your job to how you’re using AI in your job.

That’s why the 2024 UXPA Salary Survey included several new questions on previous and expected future use of AI in UX. What they found constitutes an important snapshot in time as we come to understand how AI influences what we do.

Survey Details

AI changes quickly, so knowing when our data was collected is an important qualifier. For example, image generation is now a common capability across AI platforms, when only a year ago, it was an advanced feature few had access to. The accuracy of meeting summaries and transcripts also improves each month.

The data came from respondents recruited through postings on professional networks and websites such as UXPA and LinkedIn. Additional respondents were recruited through snowball sampling. The survey ran from April through October 2024, and the final sample size was 444 respondents from around the world (67% from the U.S., 4% from the UK, 4% from Canada, and smaller percentages from 34 other countries). Most respondents reported being User Researchers (64%) and User Experience Designers (31%).

Key Findings

The three AI-related questions in the survey addressed the impact of AI on respondents’ roles in the previous year (i.e., 2023 for a survey conducted in 2024), the expected impact of AI on roles in the next year (i.e., 2025), and the expected impact of AI on the UX profession in the next five years (relative to 2024).

Impact of AI on Your Role Last Year

Respondents were asked, “How did AI impact your role in the last year? Please select all that apply.” These were the response options (results shown in Figure 1):

• Not at all

• I tried to apply some AI to my work and saw some value

• I tried to apply some AI to my work but was not impressed with the results

• Our company encouraged us to start using AI where it made sense

• Our company discouraged us from using AI

Just under half of respondents (47%) reported trying AI in the previous year (2023) and finding some value in the effort. In contrast, about a fifth of respondents (20%) reported trying AI but were not impressed with the results.

About six times as many respondents reported their companies encouraged use of AI (37%) than discouraged use (6%). More than half of respondents (57%) did not select either of these options.

Expected AI Impact on Your Role Next Year

That’s the past, but what do people think will happen?

Respondents were asked, “How do you expect AI to impact your role in the next year?” with the following response options (results shown in Figure 2):

• Not at all

• I expect to use AI less than I did last year

• I expect to use AI about the same as I did last year

• I expect to use more AI than I did last year

Just over half of respondents expect to increase their use of AI in the next year, and just over a third expect to maintain their use of AI. Only 3% expect to decrease their use of AI.

Expected Impact of AI on the UX Profession in the Next Five Years

Finally, what’s the pulse of how the field will use AI?

Respondents were asked, “How do you feel AI will impact the UX profession in the next 5 years? Please select all that apply.” These were the response options (results shown in Figure 3):

• Not at all

• I think the field will develop a better understanding of when AI is appropriate in UX design and research

• I think increased use of AI will lead to better user experiences

• I’m concerned that increased use of AI will lead to poorer user experiences

• I’m concerned increased use of AI will cause job losses in the fields of user experience design and research

Most respondents (65%) were optimistic about UX practitioners arriving at a better understanding of when AI is appropriate in UX design and research, but less than half of that percentage (30%) anticipated that AI use would improve user experiences. In contrast, 37% expected AI to lead to poorer user experiences. About a quarter of respondents (28%) expressed concern that increased use of AI would cause job losses in UX research and design.

There’s a lot more research to do on the appropriate applications of AI in UX, but benchmarking attitudes for moments in time helps us all understand how technologies may or may not change how we do our jobs or what jobs we’ll do.

Oh, and in case you were wondering if this article was written by ChatGPT or Gemini? Nope, it was written entirely by JimandI.

Summary and Discussion

The responses of 444 respondents to three questions in the 2024 UXPA Salary Survey about past and expected future use of AI in UX design and research revealed a mix of optimistic and pessimistic views.

The usage of AI in the previous year (2023) had a healthy start. Almost half of respondents had tried AI and found some value in their effort. About 20% reported trying AI but not being impressed.

More companies supported using AI than discouraged it. About six times as many respondents reported that their companies were encouraging the use of AI (37%) than discouraging its use (6%).

Most respondents expected to use AI more in the next year. In the next year, 53% of respondents expected to use more AI, 36% expected to stay about the same, and 3% expected to reduce their usage of AI.

Expectations over the next five years were mixed. There was a general belief that there would be a better understanding of how to use AI in UX design and research in the next five years (65%), and some respondents anticipated that AI would improve the user experience (30%). On the other hand, over a third of respondents (37%) thought AI would degrade the user experience, and about a quarter thought AI would cause UX job losses.