# Why the same AI course works for some of your people and not others > Discover how employees learn AI through Sam Wong's framework. Explore four axes and 16 combinations to tailor training for your team. - URL: https://hyperfocusam.com/blog/how-employees-learn-ai/ - Language: en - Author: Sam Wong (黃力桐), https://hyperfocusam.com/ - 中文版: https://hyperfocusam.com/blog/how-employees-learn-ai-tc/ (Markdown: https://hyperfocusam.com/blog/how-employees-learn-ai-tc/index.md) - Published: 2026-10-30; revised 2026-10-30 --- **Short answer:** people learn AI along different routes. In AI Learner Types, a framework I have been developing, four questions decide the route: build your own system or use a proven one, go deep or wide, act first or understand first, create or analyse. One identical course suits some of those combinations and misses the rest. *Sam Wong (黃力桐) is a corporate AI trainer based in Hong Kong. 11,000+ people trained, 70+ organisations, 13 countries.* ## Why does learning AI feel so messy? If AI training has left your team feeling confused, that is normal. AI has a long history, but using it day to day, in business or in your own life, is still very new. I recorded my thinking on this in October 2026 and put it this way: > "AI has probably been around a long time, but at the level of actually using it, it is very new, for business use and personal use alike. So it's completely normal that people feel confused." A lot of the time there is another feeling too: people sit through an AI course and come away thinking that what they learned has nothing to do with them. Part of the reason, I think, is that everyone is given the same route. ## What are AI Learner Types? You have probably heard people describe themselves by one of the 16 personality types. Some people call the 16 personality types pseudo-science, and I take that seriously. But the idea does help quite a few people see where they stand and find their own path. When I applied the same idea to learning AI, it fitted better than I expected. AI Learner Types borrows the shape: four axes, two sides each, 16 combinations. It is not a personality test. It describes what you want to do with AI right now and the order you like to learn in, and your type can change as you grow. ## What are the four axes? ### 1. Builder or User: how you want to work with AI Builders want their own system. Give them a starting point and they will explore and shape their own workflow. In my words from the recording: "You don't need to hold my hand. I'll develop my own system. I don't want to be bound by fixed rules." Users trust what has already been proven. A tested tool or workflow feels steadier to them and is more likely to get the result they want. ### 2. Deep or Wide: your knowledge orientation Deep learners (specialists) go deep in one field and apply AI to that field's problems. Wide learners (generalists) see patterns across fields and spot where AI can link several of them into one complete workflow. ### 3. Practical or Theoretical: how you bring AI into your work Practical people value frameworks. Once they understand the framework, they put it to work straight away and fill in the details later. Theoretical people want the details first, how the tool operates and the principles behind it, and only then apply it widely. ### 4. Artistic or Logical: what you are drawn to make Artistic learners are pulled towards creating something from nothing: written content, images, media. Logical learners are pulled towards data, analysis and logic, the harder side of AI use. ## How can someone tell which type they are? I start with one moment: the first few minutes with a new AI tool. My own first move with a new AI tool is usually to look for someone to show me how it works. Two signals help. **Where do you go first?** Some people open the official manual. They see roughly what the tool can do, look at what its templates can do, then try it themselves. Others look for what already exists: online tutorials, explainers on social media, written guides. This signal belongs to the Practical or Theoretical axis. I have not yet settled which first move points to which side. **Once you know what a tool is for, what is your first thought?** "How do I combine it with my other tools into one complete workflow?" points to a Builder. "What is this really worth for my own work, and how far can I take it?" points to a User. If you want to find your own type, there is a free 5-minute test at [AI Learner Types](https://hyperfocusam.com/en/ai-learner-types/). It needs no email. ## What does this mean for HR, L&D and team leads? This is the core idea, as I said it in the recording: > "So the same so-called way of learning AI gives them different results, very different both in what they learn and in what they can apply." A few practical moves follow from that: 1. **Ask before you plan.** Put one question to your people: "When you get a new AI tool, what's the first thing you do?" The answers show who wants a manual and who wants a walkthrough. 2. **Give Builders a starting point, not a script.** They want room to shape their own system. 3. **Give Users a proven workflow they can trust.** For them, stability is the point. 4. **Order the session for both sides.** Practical people need the framework first. Theoretical people need the details first. 5. **Pick examples people care about.** Content and media for artistic learners, data and analysis for logical ones. Running a programme department by department helps with this. In one [department-by-department programme](https://hyperfocusam.com/en/cases/garden-ai-pioneer-programme/), each cohort practised on exercises built around its own work. If your concern is what happens after the session, I wrote about [why AI training doesn't stick in large organisations](https://hyperfocusam.com/blog/why-ai-training-doesnt-stick/). For how I run training for teams, see [corporate AI training in Hong Kong](https://hyperfocusam.com/corporate-ai-training-hong-kong/) ([中文](https://hyperfocusam.com/zh/corporate-ai-training-hong-kong/)). ## Is this backed by science? No, and I don't want to dress it up as if it were. The 16-type idea it borrows from is contested, and AI Learner Types is a practical lens that comes from working with many different people who are learning AI. It is still a first version. Use it to start a better conversation about how your people learn. Don't use it to label them. ## FAQ ### What are AI Learner Types? AI Learner Types is a framework by Hong Kong AI trainer Sam Wong, first recorded in October 2026. It categorises how people learn AI along four axes: Builder or User, Deep or Wide, Practical or Theoretical, and Artistic or Logical. These axes create 16 combinations. While it borrows the structure of personality types, it is not a personality test, and an individual's type can change as they grow. ### Why doesn't one AI course work for the whole team? People take different routes into AI. A Builder wants a starting point and room to create their own workflow, while a User wants a proven workflow they can trust. Practical learners want the framework first, whereas theoretical learners want the details first. Sam Wong's core point is that the same learning method produces different results for different people in both learning and application. ### What is the difference between a Builder and a User? Once they understand what an AI tool is for, a Builder's first thought is how to combine it with other tools into one complete workflow. Builders dislike hand-holding and fixed rules. A User focuses on what the tool is worth for their own work and trusts approaches others have already proven, as this is steadier and more likely to deliver the desired result. ### Is AI Learner Types scientific? No. It is a practical lens, not a validated psychometric test. Sam Wong acknowledges that some people call the 16 personality types pseudo-science. He uses the shape because it helps people see where they stand and find their own path with AI. It should be used to plan better conversations and training routes, not to label or select people. ### How can a team lead use AI Learner Types? Start by asking each person: 'When you get a new AI tool, what's the first thing you do?' Use those answers to plan the session. Give Builders a starting point and freedom, while giving Users proven workflows. Lead with the framework for practical people and details for theoretical ones, choosing examples that match what your team cares about. ### How do I find my own AI learner type? Notice your first move with a new AI tool. Do you open the official manual to try it yourself, or do you look for tutorials and explainers? Once you know what the tool does, do you think about connecting it to other tools or its value for your own work? A free 5-minute test is available at hyperfocusam.com/en/ai-learner-types/ without requiring an email. --- **Planning AI training for a team with mixed learners?** [Message me on WhatsApp](https://wa.me/85264315177?text=Hi%20Sam%2C%20I%20read%20your%20article%20about%20how%20employees%20learn%20AI) and tell me a little about your people. ## Sources - Sam Wong, voice notes on AI learner types, recorded 28 October 2026 (in Cantonese; quotes translated). - The free AI learner types test on this site: [/en/ai-learner-types/](https://hyperfocusam.com/en/ai-learner-types/) (read 29 October 2026). - A department-by-department programme: [/en/cases/garden-ai-pioneer-programme/](https://hyperfocusam.com/en/cases/garden-ai-pioneer-programme/) (read 29 October 2026).