{"id":4033,"date":"2025-11-25T13:25:45","date_gmt":"2025-11-25T12:25:45","guid":{"rendered":"https:\/\/leapfrog.nl\/blog\/?p=4033"},"modified":"2025-11-25T13:25:45","modified_gmt":"2025-11-25T12:25:45","slug":"reclaiming-autonomy-designing-ai-enhanced-work-tools-that-empower-users","status":"publish","type":"post","link":"https:\/\/leapfrog.nl\/blog\/archives\/2025\/11\/25\/reclaiming-autonomy-designing-ai-enhanced-work-tools-that-empower-users\/","title":{"rendered":"Reclaiming Autonomy: Designing AI-Enhanced Work Tools That Empower Users"},"content":{"rendered":"\n

Based on an invited talk delivered at Enterprise UX<\/a>, on November 21, 2025 in Amersfoort, the Netherlands.<\/em><\/p>\n\n\n

\"\"<\/figure>\n\n\n

In a previous life, I was a practicing designer. These days I’m a postdoc at TU Delft, researching something called Contestable AI. Today I want to explore how we can design AI work tools that preserve worker autonomy\u2014focusing specifically on large language models and knowledge work.<\/p>\n\n\n

The Meeting We’ve All Been In<\/h2>\n\n\n
\"\"<\/figure>\n\n\n

Who’s been in this meeting? Your CEO saw a demo, and now your PM is asking you to build some kind of AI feature into your product.<\/p>\n\n\n

This is very much like Office Space<\/em>: decisions about tools and automation are being made top-down, without consulting the people who actually do the work.<\/p>\n\n\n

What I want to explore are the questions you should be asking before you go off and build that thing. Because we shouldn’t just be asking “can we build it?” but also “should we build it?” And if so, how do we build it in a way that empowers workers rather than diminishes them?<\/p>\n\n\n

Part 1: Reality Check<\/h2>\n\n\n

What We’re Actually Building<\/h3>\n\n\n
\"\"<\/figure>\n\n\n

Large language models can be thought of as databases containing programs for transforming text (Chollet, 2022)<\/a>. When we prompt, we’re querying that database.<\/p>\n\n\n

The simpler precursors of LLMs would let you take the word “king” and ask it to make it female, outputting “queen.” Now, language models work similarly but can do much more complex transformations\u2014give it a poem, ask it to write in the style of Shakespeare, and it outputs a transformed poem.<\/p>\n\n\n

The key point: they are sophisticated text transformation machines. They are not magic. Understanding this helps us design better.<\/p>\n\n\n

Three Assumptions to Challenge<\/h3>\n\n\n

Before adding AI, we should challenge three things:<\/p>\n\n\n

    \n
  1. Functionality<\/strong>: Does it actually work?<\/li>\n\n\n\n
  2. Power<\/strong>: Who really benefits?<\/li>\n\n\n\n
  3. Practice<\/strong>: What skills or processes are transformed?<\/li>\n<\/ol>\n\n\n

    1. Functionality: Does It Work?<\/h4>\n\n\n
    \"\"<\/figure>\n\n\n

    One problem with AI projects is that functionality is often assumed instead of demonstrated (Raji et al., 2022<\/a>). And historically, service sector automation has not led to expected productivity gains (Benanav, 2020<\/a>).<\/p>\n\n\n

    What this means: don’t just trust the demo. Demand evidence in your actual context. Ask for them to show it working in production, not a prototype.<\/p>\n\n\n

    2. Power: Who Benefits?<\/h4>\n\n\n
    \"\"<\/figure>\n\n\n

    Current AI developments seem to favor employers over workers. Because of this, some have started taking inspiration from the Luddites (Merchant, 2023<\/a>).<\/p>\n\n\n

    It’s a common misconception that Luddites hated technology. They hated losing control over their craft. They smashed frames operated by unskilled workers that undercut skilled craftspeople (Sabie et al., 2023<\/a>).<\/p>\n\n\n

    What we should be asking: who gains power, and who loses it? This isn’t about being anti-technology. It’s about being pro-empowerment.<\/p>\n\n\n

    3. Practice: What Changes?<\/h4>\n\n\n
    \"\"<\/figure>\n\n\n

    AI-enabled work tools can have second-order effects on work practices. Automation breaks skill transmission from experts to novices (Beane, 2024<\/a>). For example, surgical robots that can be remotely operated by expert surgeons mean junior surgeons don’t learn by doing.<\/p>\n\n\n

    Some work that is challenging, complex, and requires human connection should be preserved so that learning can happen.<\/p>\n\n\n

    On the other hand, before we automate a task, we should ask whether a process should exist at all. Otherwise, we may be simply reifying bureaucracy. As Michael Hammer put it: “don’t automate, obliterate” (1990<\/a>).<\/p>\n\n\n

    Every automation project is an opportunity to liberate skilled professionals from bureaucracy.<\/p>\n\n\n

    Part 2: Control \u2192 Autonomy<\/h2>\n\n\n

    All three questions are really about control. Control over whether tools serve you. Control over developing expertise. This is fundamentally about autonomy.<\/p>\n\n\n

    What Autonomy Is<\/h3>\n\n\n
    \"\"<\/figure>\n\n\n

    A common definition of autonomy is the effective capacity for self-governance (Prunkl, 2022<\/a>). It consists of two dimensions:<\/p>\n\n\n