UX Roundup: Jakob Keynote Video | University Courses | 8 Uses of AI for UX Design | Museum Usability
Summary: Full video from my Oslo keynote now online | Promising university courses | Main uses of AI in the UX design process | Museum usability
UX Roundup for November 18, 2024. (Ideogram)
Full Video from My Oslo Keynote Now Online
The full video from my keynote at the Y Oslo conference is now online (YouTube, 59 min. video).
Last month’s Y Oslo was a great conference, with 1,400 Norwegian UX freaks filling the Oslo Concert Hall. I started my talk with a historical comparison with the UX field when I started in 1983: at that time, the entire world had fewer UX professionals than were present in that one room in Norway for my talk in 2024. We’ve come a long way!
At the tradeshow part of the event, I was intrigued to see several Norwegian companies presenting products targeting the UX industry, often using AI.
Most noteworthy was Breyta which has an AI product to summarize information across large sets of files, such as transcriptions of discovery research sessions. One use case they touted was the ability to ask questions of your UXR report repository, bringing life to formerly dusty report collections. For example, find out “what user behaviors we have seen in past studies that relate to topic X.” This might save companies from running repetitive studies when they don’t know what they already have found in past research.
Jakob Nielsen keynoting the Y Oslo conference, followed by Q&A. The Oslo Concert Hall was full, as were the networking sessions. Meeting people in person is much more exciting than attending online events. Great crowd! (Photos courtesy of Y Oslo)
Breyta had a demo where the document collection was created in real time from transcripts of the conference talks. Before my own talk, I asked the AI about the biggest disagreement between the speakers so far. I had hoped to identify a controversial issue I could comment on, but the AI claimed that the speakers agreed on all major issues. This sounds credible because traditional UX is a settled field now that we have moved on to new grounds with AI.
In the hope of creating controversy, I mentioned that good vice presidents make bad presidents because they tend to change the company from “founder mode” to “manager mode.” But I think people have seen this movie too often for my point to be that controversial: the founder leaves, the most promising tier-two manager takes over, and the company promptly loses its vision and starts to decline. [See video clip of this segment of my keynote.]
Even though that one statement may not have been as controversial as I had hoped, I still made plenty of statements in the session to stir the juices. It’s the job of a keynote speaker to get the audience to think outside their everyday job. To give you a summary, I asked Claude to identify 10 controversial things I said and explain why they are notable:
AI’s Impact vs Previous Tech Revolutions: AI is bigger than PC, Web, and Mobile revolutions combined; comparable to Agricultural and Industrial revolutions. [See article about the AI revolution compared with PC and Web.]
Notable because: This is a bold claim that positions AI as a civilization-changing force rather than just another tech advancement.
The “Mediocrity” Achievement: My career has been a pursuit of mediocrity — moving UX from “misery” to “mediocre.” [See video clip about mediocre design (0:52 min.)]
Notable because: It’s a provocative and humble take that challenges the typical narrative of technological progress, suggesting we’re still far from where we should be.
AI Intelligence Progression Timeline: AI will progress from high school level (current) to university level (2025) to PhD level (2027) to superhuman (2030). [See article about the AI scalability law.]
Notable because: This is a specific and potentially controversial timeline that makes bold predictions about AI capabilities.
The human brain can only get so large, limiting meatware intelligence, even in top scientists with big heads. AI suffers no such limitations, and there’s no end to how large training clusters we can build. The AI scaling law predicts that AI superintelligence will exceed anything seen in even the most brilliant humans. Coming around 2030. (Midjourney)
AI and Employment: AI won’t cause unemployment in UX field; instead it will enable much more work to be done. [See article about unemployment.]
Notable because: This counters common fears about AI replacing jobs, suggesting instead a transformation of work.
AI’s Superior Empathy: Studies show AI demonstrates higher empathy than humans in certain contexts, like medical consultations. [See article about AI empathy research.]
Notable because: This challenges fundamental assumptions about human uniqueness in emotional intelligence.
The “Pancaking” of UX Departments: AI efficiency will lead to flatter organizational structures with smaller, more efficient teams. [See article about pancaking of UX.]
Notable because: This predicts a fundamental restructuring of how companies operate.
Organizations will become flat as a pancake as hyper-efficient small teams ship products with AI. Forget any career plans of climbing the corporate ladder because there will be hardly any higher levels of management left. (Midjourney)
Critique of Academia: Universities and academia have failed UX practitioners by focusing on esoteric research instead of practical applications. [See video on how academic research failed UX (3:07 min.)]
Notable because: This is a direct criticism of the academic establishment in the UX/HCI field.
AI's Impact on Skill Gaps: AI narrows skill gaps by helping less skilled people more than highly skilled ones.
Notable because: This contradicts common assumptions about AI primarily benefiting the most skilled workers.
AI is a forklift for the mind: it helps users lift heavy cognitive burdens. Just like a real forklift helps weak warehouse workers more than strong ones in terms of maneuvering heavy loads, AI has proven to help the lower end of knowledge workers even more than it helps the best knowledge workers. (Midjourney)
The End of Universities: We might not have human teachers for adult learners in 20 years due to AI’s superior teaching capabilities. Full-day professional-development courses will die in 2-3 years as AI just-in-time training becomes the way to learn focused topics, 12-week UX bootcamps may last 4-5 years because it’ll take a little longer for AI to become superior at teaching a full set of practical skills. Universities might keep going for as much as 20 years due to the intense conservatism of academia.
Notable because: This is an extremely controversial prediction about the future of education.
Traditional UX Research Validity: Many 25-year-old UX guidelines are still valid despite massive technological changes. [See article about the history of the 10 usability heuristics.]
Notable because: This suggests fundamental principles of human-computer interaction remain constant despite technological advancement.
If you don’t want to watch my full one-hour keynote session (including interesting Q&A), here are a few shorter clips:
You are not the user (0:22 min.)
The 4 metaphors for working with AI (1:51 min.)
AI is more significant than the PC, Web, and mobile tech revolutions put together (2:08 min.)
Promising University Courses
Despite my critique of academia as being mainly useless for practical UX (see above news item), there are honorable exceptions. The City University of London St George’s has a master’s degree in user experience with a great set of courses. (They are currently taking applications for the 15-month program that starts in January — if you’re not ready, there will be another chance in January 2026.)
I’m particularly pleased that 2 out of the 6 required “core” courses (i.e., 33%) are about AI and UX:
Experiencing AI: how to design AI-based products or integrating AI elements into existing products.
AI for Designers: how to use AI tools and capabilities throughout the UX design life cycle, with a focus on using AI for innovation and creativity.
These are exactly the two aspects of AI that UX professionals need to know. As with most graduate programs, City University also includes a dissertation project. If you attend this program, I encourage you to choose an area of AI-UX for your thesis research. There are so many unstudied aspects of AI, both regarding its use in user interfaces and its use in the design project. You can’t help but do pioneering work if you pick a topic in this realm.
(Hat tip to Professor Nick Hine for alerting me to this exciting upgrade to the City University MSc.)
City University of London now teaches not just one, but two, courses on AI in UX as part of its master’s degree program. (Midjourney)
Main Uses of AI in the UX Design Process
After my talk at Y Oslo, André Fangueiro (Head of Design at Tietoevry) gave a fascinating talk about how he and his team use AI in their design process (YouTube, 20 min. video).
Here are 8 key uses he discussed in his talk:
1. Research Repository Analysis
What: AI-powered platform that centralizes and analyzes past user interviews and research
Why: Prevents redundant research, leverages existing insights
Special benefit: Can search through video/audio content in multiple languages and synthesize relevant insights quickly
Without AI, your archive of old user research reports becomes a dead-end storage from which no past insights are ever retrieved. (Midjourney)
2. Workshop Planning and Organization
What: AI assistance in organizing workshops, creating icebreakers, and structuring activities
Why: Reduces workshop preparation time from hours to minutes
Special benefit: Can analyze meeting transcripts to organize relevant workshop content and extract key ideas automatically
3. Rapid Prototyping
What: AI-powered code generation for quick website/interface prototypes
Why: Enables real-time prototype creation during client meetings
Special benefit: Allows immediate visualization and iteration of ideas with working code
4. Concept Validation
What: AI agents with different expertise personas reviewing design concepts
Why: Enables rapid iteration and validation of thousands of concepts
Special benefit: Can process 6,000+ iterations in minutes, providing multiple expert perspectives
5. Automated Visualization
What: AI generation of visual representations of concepts
Why: Quickly creates visual assets for testing
Special benefit: Can be directly linked to social media for immediate user feedback
6. Customer Feedback Loop
What: AI system that integrates real human feedback with AI analysis
Why: Creates continuous improvement cycle
Special benefit: Combines machine learning with actual user responses to improve future iterations
7. Workshop Content Analysis
What: AI analysis of meeting transcripts to extract key points and ideas
Why: Ensures important discussion points aren’t lost
Special benefit: Automatically organizes and categorizes discussion content
8. Project Brief Analysis
What: AI evaluation and summary of project briefs
Why: Quickly identifies key requirements and objectives
Special benefit: Provides instant justification and validation of brief understanding
What happened in this design workshop? AI can give you a summary and ensure you don’t forget action items. (Midjourney)
André Fangueiro emphasized that AI tools should be used to:
Free designers from mundane tasks
Increase speed of iteration
Enable focus on higher-value activities
Achieve approximately 80% productivity gains
Enable faster customer-validated innovation
How many of these 8 valuable ways of AI empowering your UX team do you employ right now? If there’s even one you don’t use, experiment with it on your next project, and you’ll have earned back your subscription fee for this newsletter many times. (Since I charge $0, this is an easy promise, but even if subscriptions cost $100/year, I still think you’d be ahead, just from this one story.)
Museum Usability: Oslo Delivers
The first painting you see when visiting the Norwegian National Gallery in Oslo is the Mona Lisa. Wait a minute, I thought that painting was in the Louvre in Paris? Yes, the original painting is in the Louvre, but the Norwegian museum has a copy.
When the museum was founded in 1836, its first board felt the need to import some European culture to then-barbarian Norway. They thus bought what they believed to be a replica of the Mona Lisa painted in Leonardo da Vinci’s studio by one of his students. (They did know that they weren’t getting the original.) However, modern research has shown that the Norwegian canvas was painted quite some time after Leonardo and has no connection to the artist or his students.
What’s nice about this story is that the museum admits its mistake and keeps the painting on display, with a corrected label. Anybody can make a mistake — smart people own up and learn.
I was very impressed with the Norwegian National Gallery, which I visited in connection with my speech at the Y Oslo conference. The collection is great, but so is the visitor experience. So many museums have terrible usability, with a large percentage of modern museum redesigns featuring labels with small, thin gray fonts on an only slightly-lighter gray background. You would think that museums hire people who have been fired as web designers after designing websites with no sales because the potential customers couldn’t read the text.
It's also very common to get lost in a labyrinth of galleries with no guidance for how to progress between the rooms.
The Norwegian National Gallery delivers on both accounts: (reasonably) readable labels and easy navigation between rooms. Each room features its gallery number in huge numerals on the wall (at least 30 cm/one foot high), and every single doorway clearly shows the numbers of the two rooms it connects. For visitors with tired feet, every room includes a bench in minimalist (but comfortable) modern Norwegian design.
Kudos to the Norwegian museum designers. Museums from the rest of the world should go on a field trip and learn.
The Norwegian National Gallery has great visitor usability and is worth a visit. The label text could have been a little larger, though. (Real photo)
About the Author
Jakob Nielsen, Ph.D., is a usability pioneer with 41 years experience in UX and the Founder of UX Tigers. He founded the discount usability movement for fast and cheap iterative design, including heuristic evaluation and the 10 usability heuristics. He formulated the eponymous Jakob’s Law of the Internet User Experience. Named “the king of usability” by Internet Magazine, “the guru of Web page usability” by The New York Times, and “the next best thing to a true time machine” by USA Today.
Previously, Dr. Nielsen was a Sun Microsystems Distinguished Engineer and a Member of Research Staff at Bell Communications Research, the branch of Bell Labs owned by the Regional Bell Operating Companies. He is the author of 8 books, including the best-selling Designing Web Usability: The Practice of Simplicity (published in 22 languages), the foundational Usability Engineering (27,753 citations in Google Scholar), and the pioneering Hypertext and Hypermedia (published two years before the Web launched).
Dr. Nielsen holds 79 United States patents, mainly on making the Internet easier to use. He received the Lifetime Achievement Award for Human–Computer Interaction Practice from ACM SIGCHI and was named a “Titan of Human Factors” by the Human Factors and Ergonomics Society.
· Subscribe to Jakob’s newsletter to get the full text of new articles emailed to you as soon as they are published.
· Read: article about Jakob Nielsen’s career in UX
· Watch: Jakob Nielsen’s 41 years in UX (8 min. video)
This was great read. 8 ways to incorporate UX was helpful. You mentioned “AI agents with different expertise personas reviewing design concept”. How do I go about doing that?
Do I train multiple GTPs based on Role?
Is it simply writing “Act as a Product Manager” in the prompt?
Thanks in advance,
Sarah Soha