Why AI? Keynote at Wales UXR Summit 2025

Keynote at the Wales UXR Summit 2025, Ty Admiral, Cardiff, 22nd October 2025

https://alandix.com/academic/talks/Wales-UXR-summit-2025-why-AI/

In some ways AI is just another technology we still need to understand users and their contexts, and match the appropriate technology with user needs even if that sometimes means not using technology at all. However, AI is different. We can think of the particular qualities of AI in terms of one A – agency – and three Cs – complexity, (un)certainty and co-adaptation. AI can be used in user facing systems, and user research needs to think about how to design for these circumstances. The second C, uncertainty, is critical here. AI can also be used as part of the user experience research and development process. There is justified fear that AI may be seen as a cheap alternative for UX professionals, but ideally AI should be used to enhance rather than replace, both in the products we create and in the processes used to create them. Let's use AI in UXR to do better rather than simply cheaper.

1. Alan Dix Why AI? https:// alandix.com /academic/talks/Wales-UXR-summit-2025-why-AI/

2. is AI special just like any technology understand users, see them in their own context match tech to needs whether to use AI at all – printer vignette but also special … A – agency three Cs complexity (un)certainty co-adaptation

3. what kind of AI traditional well defined rules in specific domains inspired by rational thinking problems with ill defined things + general knowledge (relatively) reliable and transparent (deep) neural nets good for ill defined things black-box – explainable AI DNNs – enabled by big compute and data of 2000’s web search, recommender systems +big data stats etc.!!

4. what kind of AI … LLMS started as next word predictors using vast text corpora good for vague things + general knowledge few guarantees type I / II – thinking fast and slow think Jung not Einstein Alien intelligence

5. where to use AI? AI in UX design process Gov.uk identity validation pattern hunting in log data don’t forget product managers, sales and marketing, help desks (lots of )throw away prototypes personas and diversity … empathy UX for AI in products matching AI to need image processing, free text/speech, structured data, … need for explanation? plus ….

6. it’s not about AI … appropriate intelligence: AI is flaky, imprecise, error prone – it doesn’t need to be perfect the whole system AI + human + GOC suggest (to human) then act suggest to code and test like eCommerce good old-fashioned code