Critsly: An Artefact-Aware AI Critique Teammate for Design Education and Project-Based Learning
The paper introduces Critsly, an artifact-aware AI workspace that transforms design critique from isolated chatbot feedback into a structured, collaborative teammate workflow by integrating visual boards, multi-perspective personas, and actionable reflection tools to enhance design education and project-based learning.
Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer
Imagine you're in a design studio, surrounded by sketches, sticky notes, and half-finished models. Usually, getting feedback on your work is like waiting for a rare, unpredictable lightning storm. It happens only when a teacher or a judge is available, it's often just a quick chat that vanishes into thin air, and it's hard to look back at what was said later.
Enter Critsly, a new digital workspace that tries to turn Artificial Intelligence from a distant, robotic commentator into a critique teammate sitting right next to you at your workbench.
The Problem with "Chatbot" Feedback
Most AI tools today are like a person shouting advice from a megaphone across a field. You have to shout your question (a "prompt"), and they shout back an answer. But in design, the thing you are making matters just as much as your words. A chatbot doesn't really "see" your sketchbook, your sticky notes, or the history of how you changed your mind.
Critsly argues that this "shouting from afar" approach is missing the point. Instead, it suggests that AI should be board-aware. Think of Critsly as a smart assistant who can actually see your entire workbench. It doesn't just listen to what you say; it looks at your drawings, your notes, your links, and your past changes to give feedback that actually fits your project.
How It Works: The "Reflecture" Flow
When you use Critsly, you don't just ask for "feedback." You go through a guided process called Reflecture. It's like a coach walking you through a specific routine:
- State your intent: Before the AI speaks, you tell it what you were trying to do and what your goals are. This stops the AI from giving vague advice.
- Wear different hats: This is the coolest part. Critsly lets you ask for feedback from different "personalities," inspired by the famous "Six Thinking Hats" idea. You can ask the AI to look at your work through a risk-focused lens (spotting dangers), an emotional lens (how it feels), or a generative lens (how to make it better). It's like having a whole team of experts with different specialties, all in one digital room.
- Make a plan: The AI doesn't just criticize; it helps you synthesize what it saw and suggests a concrete action plan for your next revision.
The "Teammate" vs. The "Judge"
Here is the most important thing to remember: Critsly is not a replacement for human teachers. The authors are very clear that this system is a learning environment for practice, not a final grading machine.
Think of it this way: Critsly is the sparring partner you train with between real matches. It lets you rehearse your critique skills, try out different perspectives, and turn feedback into action before you face a real jury or teacher. The human teacher is still the referee and the judge; they just get a special "evidence view" that shows them how you practiced, what "hats" you tried on, and what action plans you made. This helps the teacher see where you might need extra help, without the AI ever taking over the final decision.
What We Know (and What We Don't)
The paper presents a working system that has been built and tested in a live environment. You can actually go to their website and try the "Critique Studio" workflow. The authors have verified that the board initializes, the critique panel works, and the canvas interacts correctly.
However, the authors are careful not to claim they have proven that this makes students smarter or better designers yet. They explicitly state that this submission does not claim measured learning gains or validated automated assessment scores. Instead, they suggest that this system could make critique more frequent, more structured, and easier to inspect. They are offering a working example and a suggestion that this approach is promising, inviting others to study how it affects real learning in the future.
In short, Critsly is a playful, structured, and smart way to bring AI into the messy, creative process of design—not to do the work for you, but to be the best possible teammate you've ever had while you do it.
Drowning in papers in your field?
Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.