Read the evidence without turning it into a legal shortcut.
These explainers translate Australian and international studies into practical reading for legal, product, UX, compliance and engineering teams. Each page shows what was measured, what the authors found and where the conclusion stops. A research label does not establish the complete section 28B test or any other Australian contravention.
No matching guides. Try a broader search.
Dark-pattern types examined16
GUI agents in phase one6
Study phases2
Selected values from the reviewed study. Open the guide for the method, denominator and limits.
Peer-reviewed study
When an AI agent shops or subscribes, who notices the manipulation?
The 2026 CHI study on GUI agents and dark patterns translated for Australian pricing, subscription and cancellation journeys.
Original work
2026
Method
Six GUI agents in phase one and 16 dark-pattern types across diverse controlled tasks.
What Australia’s “Patterns in the dark” report adds to the debate
The University of South Australia’s commissioned landscape review explained: its taxonomy, Consumer Data Right context and the limits noted during official discussion.
Original work
2024
Method
Published deceptive-pattern research and regulatory taxonomies reviewed for an Australian data-sharing and consent setting.
Selected values from the reviewed study. Open the guide for the method, denominator and limits.
Institutional research
What 2,000 Australians said dark patterns cost them
The CPRC survey behind “Duped by design” explained: reported financial, privacy and trust consequences, the ten tested patterns and the limits of self-reported evidence.
Original work
2022
Method
2,000 Australian consumers, with findings reported for the full sample and selected age groups.
Selected values from the reviewed study. Open the guide for the method, denominator and limits.
Peer-reviewed study
Why the quietest manipulation may deserve the closest review
The Luguri and Strahilevitz experiments for Australian teams: strong behavioural effects, weaker backlash to mild designs and careful limits on legal translation.
Original work
2021
Method
Census-weighted samples of adult participants in the United States.
Knowing a dark pattern is there may not help people resist it
A 406-person study found that people could recognise manipulative design yet remain vulnerable or uncertain about harm, challenging “just inform users” responses.
Original work
2021
Method
406 survey participants, with analysis including age-related differences in recognition and perceived harm.
Selected values from the reviewed study. Open the guide for the method, denominator and limits.
Peer-reviewed study
Why automated dark-pattern detection still needs a journey review
The 11,000-site crawl remains a foundation for automated detection, but Australian teams need authenticated journeys, operating evidence and legal review beyond page signals.
Original work
2019
Method
Approximately 53,000 product pages on about 11,000 shopping websites.
