An interesting attempt to separate user role and expertise level in evaluating research on user-centered AI systems was made in the UMAP 2025 paper "Disentangling Stakeholder Role and Expertise in User-Centered Explainable AI," presented today by Maxwell Szymanski. Placing existing work in this two-dimensional space will help us understand what we are assessing and compare different works. https://lnkd.in/exrN9g7n
UMAP 2025 paper on user-centered AI systems
This title was summarized by AI from the post below.
More Relevant Posts
-
True advancement comes from understanding where we add value and applying AI in ways that strengthen rather than imitate. This idea reaches beyond national strategy. It applies to how every organization & leader approaches technology. Wisdom in AI is not about possession. It is about alignment between human purpose & machine capability. That balance is at the heart of our Responsible AI Webinar on 10/22. Learn more here: https://lnkd.in/eH5Vhuxt
Dr. Murthy questions the rush toward AI nationalism and asks whether every country truly needs its own language model. He argues that progress depends on understanding strengths and applying AI where it adds real value rather than following trends. His perspective reflects the kind of thoughtful approach that guides our work at Soothsayer Analytics. To learn more about how we think about responsible and practical use of AI, join our upcoming Responsible AI Webinar on October 22nd:
-
Context is a critical but finite resource for AI agents. In this post, Anthropic team explore strategies for effectively curating and managing the context that powers them. Link to the blog - https://lnkd.in/gTY6zzfm
-
-
The few things I learned from this module- 1. What is LLM and Generative AI? 2. How are the growing interests by scientists, collaborators and editors increasing with the time? 3. what are the pros and cons of using Generative AI? 4. Upto what level and how Elsevier is allowing it's writers to use AI under human observation? 5. What are the regulations are being made in China, European Union and America.
-
One thing I deeply love about the emerging AI world is that it consistently brings us back to philosophy. In the brilliant research paper conducted by Philipp Koralus he dives deeper into focusing more on AI's that ask better questions using the Socratic method... Not one's that just word vomit and answer to you. Read the full paper in the link below: https://lnkd.in/dRKuV5Tf
-
-
A thought provoking article on the value of time and reimagining when and how we perform our work. This consideration is particularly relevant with the implementation of AI and accelerated changes in work activities for both time and space. https://lnkd.in/eCpyCcqD
-
Closing out September with a new Scientific American graphic, created for a story by Rachel Nuwer and edited by Allison Parshall. The story highlights new research published in Nature showing that humans are more likely to cheat when using AI. The Marimekko chart illustrates how participants were more inclined to act dishonestly for profit under AI-involved conditions, often by reporting a die-roll result that was higher than the observed roll. The study compared different scenarios, including a supervised learning condition where the AI was trained on biased or unbiased data, and a goal-based condition where the AI was given a high-level goal to prioritize either profit or honesty. Check out the story 👉 https://lnkd.in/gZVQjqty
-
-
AI is amplifying our cognitive biases in everything we do. The problem is, we are unaware of our biases even without AI. Now, we have a much bigger issue. In my latest newsletter issue, I talked about why we shouldn't rely too heavily on the results of AI. The post is available on my blog as well as on my Substack.
