Hi all,
I have worked on the web since 1994, and I have seen a tremendous amount of change. Of course, some of those changes were gradual, but there was also the occasional watershed moment where things suddenly shifted.
The publishing of Designing with Web Standards by Jeffrey Zeldman springs to mind. The release of the iPhone and Steve Jobs’s refusal to support Adobe Flash is another. Then, of course, Ethan Marcotte’s article led to the explosion of responsive design. I could go on, but you get the point.
Over the last few months, I have felt that same sense that things are about to shift, but this time because of the massive advances we are seeing in AI (artificial intelligence).
AI has only had a limited impact on us, but that is about to change
As is often the way with these things, there has been a lot of talk about AI. However, apart from a few notable exceptions, few organizations are integrating AI into their marketing or UX operations in any meaningful way.
Unless you work at a tech company or on a large ecommerce site, you probably haven’t felt the impact of AI that much. However, that is beginning to change, and I suspect we will find ourselves at one of those rare tipping points where we will witness a significant shift in 2023.
How I am using AI today
But instead of guessing what the future might hold (something plenty of others has done – see this and this as examples), I want to share with you how I have started using AI to help in my work today.
ChatGPT
Probably the most used AI tool in my arsenal is Chat GPT. You may have read a bit about it online, but unless you have used it regularly, you may not have realized its full potential.
I use Chat GPT to brainstorm ideas, write document outlines, compose tweets, make my emails friendlier, and much more. Recently, I even used it to suggest a possible information architecture based on the site content I provided.
I have even been using it to bug-fix my code and provide me with reference material to support business cases I have to write.
I decided to ask Chat GPT how it could help a UX designer, and this is what it answered:
- Generate ideas for new design concepts by asking it to suggest possible solutions to specific design challenges.
- Quickly prototype conversational interfaces by asking ChatGPT to generate sample dialogue for a chatbot or voice assistant.
- Generate copy for a website or app by asking ChatGPT to write descriptions of features, product benefits, and other content.
- Conduct user research by asking ChatGPT to generate sample questions for user interviews or surveys.
If you haven’t tried Chat GPT, I highly recommend doing so. I have found it enormously helpful and cannot wait to see similar functionality integrated into Notion (another favorite tool of mine).
DALLE 2 and Midjourney
OpenAI, the company behind Chat GPT, is also the creator of DALL·E 2, which can generate imagery based on text prompts.
Although DALL·E 2 is great fun to play with, it isn’t something I use as much daily. The images can be a bit hit-and-miss, so they aren’t generally good enough to use on a website yet. However, they can be a good source of inspiration in terms of imagery and color palettes.
Midjourney, on the other hand, is stunning, and I am constantly amazed at what it produces. I can see myself using this as a source of imagery for websites moving forward.
Khroma
Talking of color palettes, I have discovered a new tool called khroma. Khroma uses AI to learn your color preferences and then creates color combinations based on your choices, color theory, and accessibility. I haven’t used this on a project yet, but I could see it as a good way of engaging stakeholders in picking colors while avoiding ending up with something hideous.
Attention Insights
One AI tool I do use regularly is Attention Insights. That has taken thousands of hours of eye tracking studies and uses AI to predict with a high degree of accuracy (90-94%) where people will look on a page.
It is a sanity checker for me as a designer and a great way of getting clients off your back if they think people will miss some page element.
The future of AI in user research
One area I am excited about AI’s potential is user research. Wading through recordings of user interviews and usability testing videos can prove painful. AI has the potential to cut through all of that by providing aggregated data from those sessions.
I have experienced this a little when using Maze, whose excellent reports summarize test sessions.
But, the one tool that promotes its AI capabilities is Wevo. I haven’t tried it yet, but I have been given a demo, which is incredibly impressive. Its ability to analyze a large number of open response surveys and understand sentiment will make it an invaluable tool in my conversion optimization work.
However, beyond those two examples, I haven’t seen many AI tools targeted at UX professionals. Perhaps there are, and I have just missed them. If so, please let me know.
That said, I am confident that is all about to change as AI hits the mainstream.
Everything is about to change
If like me, you have worked in our industry for any length of time, you have seen supposed revolutions fizzle out. You begin to understand what will impact the web more widely and what will become a niche or disappear entirely. AI feels like the former to me. It feels significant, and the applications are wide-ranging.
My advice is to keep an eye on AI. Pay attention to applications that use AI technology for tasks you do. AI won’t take your job, but it could make it a lot easier, and those who adopt these technologies early will be well-positioned for the future.
