This summary was generated using automated tools and was not authored or reviewed by the article's author(s). It is provided to support discovery, help readers assess relevance, and assist readers from adjacent research areas in understanding the work. It is intended to complement the author-supplied abstract, which remains the primary summary of the paper. The full article remains the authoritative version of record. Click here to learn more.
AI Sycophancy and Decisions
We examine whether sycophantic AI advice distorts decisions. Our experiment involves 1,500 participants in 30 decision environments spanning core domains in eco
The New Science of Designing for Humans (SSIR)
The rise of behavioral science and impact evaluation has created a new way for engineering programs and human interactions. <meta property=
Mental Models: The Best Way to Make Intelligent Decisions (~100 Models Explained)
The smartest people in the world use mental models to make intelligent decisions, avoid stupidity, and increase productivity. Let's take a look at how ...
Cognitive Bias Lab | Learn to Make Better Decisions
Explore cognitive biases with interactive tests, simulations, and real-world examples. Free platform to sharpen decision-making and critical thinking — no sign-up needed.
Abeba Birhane on Twitter / X
a classic case of “the human mind is afforded less complexity than is owed, and the computer is afforded more wisdom than is due.” Baria and Cross (2021)that "vibe" is at the core of what makes us human. it defies formalization and datafication https://t.co/AXELAqlYcq— Abeba Birhane (@Abebab) February 17, 2024
Why I don’t trust most human-AI interaction experimental research – Jason Collins blog
Behavioural economics, data science and artificial intelligence.
The Influence of AI on Human Decisions in DHS Surveillance
DHS avoids classifying AI tools as high-risk by arguing a human reviews the AI’s output. But this defense rests on a fallacy, writes Emily Froude.
Study: Sycophantic AI can undermine human judgment
Subjects who interacted with AI tools were more likely to think they were right, less likely to resolve conflicts.
AI, peer review and the human activity of science
When researchers cede their scientific judgement to machines, we lose something important.
Computer power and human reason : from judgment to calculation
xii, 300 pages : 24 cm; Includes bibliographical references and index
Sung Kim (@sungkim.bsky.social)
I like this quote on code review: "The idea that humans will reliably look through code to find the increasingly rare issues that automated tools miss seems like a fantasy." By Marc Brooker, AWS.
Ground Decisions
Planes don’t fly themselves – they just cruise themselves. Takeoff, landing, and the big decisions happen with humans. Same with AI. What are the ground decisions in the software factor…
ECHO: Evidence-Calibrated Human Outcomes from Synthetic Users for Interaction Design
This position paper proposes ECHO, a protocol for treating simulated users built with AI as calibrated measurement instruments for scoped interaction-design claims, not as synthetic replacements for human participants. A synthetic user is an echo, not the person who produced the original signal; it gains evidentiary force only when it supports evidence-calibrated human outcomes: claims about human-facing outcomes bounded by human data. The issue matters now because LLMs and GUI agents generate logs, rationales, personas, and usability reports at product speed, while NLP and HCI lack shared norms for when such traces count as evidence. We contribute an evidentiary standard specifying when synthetic traces may support discovery, rehearsal, pilot triage, or bounded decision support. The standard scopes a design claim, records governed human seed traces, documents simulator policy, estimates design effects with human-corrected evidence, and reports uncertainty and rejection flags. It synthesizes usability inspection, personas, user simulation, agent benchmarks, social simulation, validity, correction, documentation, privacy, and human-AI interaction. The central position is narrow: synthetic users can help interaction design when they improve issue discovery, pilot triage, or bounded decision support under calibration. Without calibration, fluent traces turn model plausibility into counterfeit user evidence.
A rational analysis of the selection task as optimal data selection.
A rational analysis of the selection task as optimal data selection.
