DesignPref: Why One-Size-Fits-All AI Design Judges Miss the Mark

Kari Jaaskelainen

26 Nov 2025 — 1 min read

TL;DR

People don’t agree on what ‘good’ UI design looks like—and AI shouldn’t assume they do.

DesignPref is a new dataset: 12,000 head-to-head UI design comparisons, each labeled by 20 professional designers with multi-level ratings and short rationales. Designers often disagreed (low agreement), driven by different priorities like layout, typography, color, hierarchy, and personal style.

Why it matters: Training AI judges by majority vote smooths away real, individual taste. The authors show that personalized models—fine-tuned or prompted with designer-specific notes (RAG)—predict a designer’s choices better than ‘average’ models, even with 20× fewer examples.

First dataset for personalized visual design evaluation
Challenges the ‘one objective score’ myth in design
Opens the door to AI that adapts to your taste

Paper: https://arxiv.org/abs/2511.20513

Paper: https://arxiv.org/abs/2511.20513v1

Register: https://www.AiFeta.com

#AI #UXDesign #UIDesign #GenerativeAI #Personalization #HCI #DesignResearch #Datasets

Meet GRACE: a moral governor for safer, more transparent AI

AI agents are getting powerful—so how do we make sure they do the right thing, not just the effective thing? Meet GRACE, a reason-based moral governor that keeps AI behavior aligned with human norms by separating moral reasoning from goal-driven decision-making. * Moral Module: uses deontic logic and explicit reasons

Why some fine-tuned LLMs miss phishing—and how to fix it

Not all fine-tuned LLMs spot phishing equally. A new study tests Llama 3.1 8B, Gemma 2 9B, and Mistral on high-stakes phishing detection—and uses SHAP and mechanistic interpretability to reveal why models do (or don’t) generalize. * Architecture × data diversity matters: Gemma 2 9B hits state-of-the-art performance (F1

AI that explains itself—by following the science

AI that explains itself—by following the science Black-box AI can be powerful, but it often can’t tell us why it made a decision. Concept Bottleneck Models (CBMs) try to fix that by predicting human-understandable concepts first, then the final answer. The catch: standard CBMs ignore domain-specific cause-and-effect and

Meet GenomAgent: A Team of AI Specialists for Smarter Genomics Q&A

TL;DR: Finding reliable answers in genomics is hard. GenomAgent turns one big AI into a coordinating team of specialists—and beats the current leader by 12% on a key benchmark. Genomic facts live across many databases. Standard chatbots struggle because they can’t flexibly query those sources. GeneGPT added

TL;DR

Read more

Meet GRACE: a moral governor for safer, more transparent AI

Why some fine-tuned LLMs miss phishing—and how to fix it

AI that explains itself—by following the science

Meet GenomAgent: A Team of AI Specialists for Smarter Genomics Q&A