Outsmarting the Smart: New AI Tools Empower Users Against Digital Deception

Notre Dame researchers develop AI tools, Privacy Sandbox and Dark Pita, to enhance digital literacy and empower users to understand and control data use on online platforms.

In an era where digital savvy is becoming almost as crucial as literacy, researchers at the University of Notre Dame have found a want to us AI to bolster digital literacy and empower users navigating the (often deliberately) tricky terrain of online platforms.

The team, led by Toby Li, Assistant Professor of Computer Science and Engineering at Notre Dame, introduced two browser plug-ins: Privacy Sandbox and Dark Pita. These tools are designed to help users understand and manage how their personal data is manipulated and utilized online.

Privacy Sandbox allows users to explore and modify their online privacy settings in a risk-free environment. By using personas generated by OpenAI’s GPT-4, the plug-in simulates user interactions on various websites, revealing how personal attributes like age, race, and location influence the way platforms target content and ads. This transparency aims to provide users, a clearer view of the often opaque privacy practices on digital platforms.

The Dark Pita browser plug-in detects dark pattern designs, notifies the user and allows the user to customize their online experience. Credit: Toby Li / University of Notre Dame

Dark Pita, on the other hand, tackles the issue of “dark patterns”—subtle design tactics used to manipulate user actions on websites. This plug-in identifies such patterns and informs users of their implications, such as financial loss or privacy breaches. It also offers tools to modify the site’s code, allowing users to neutralize these manipulative elements.

Researchers who rely on digital platforms for data gathering, survey distribution, or behavioural studies can use these tools to gain a deeper understanding of how online environments influence user behaviour and decision-making. This can lead to more robust research methodologies and insights into human-computer interaction, especially in studies related to consumer behaviour, psychology, and social sciences.

This research highlight a shift towards what is sometimes called “democratizing” AI technology. By making such tools available to the public, researchers at Notre Dame are attempting to level the playing field between consumers and corporations but are also setting a precedent for the ethical use of AI.

As these tools become more widely available, they could serve as essential resources for researchers interested in exploring the intersections of technology, ethics, and society. This could catalyse further research into effective ways to bridge the gap between AI capabilities and user control, ensuring that future digital environments foster fairness, transparency, and user empowerment.

You can read both the studies at the links below.

Chaoran Chen et al, An Empathy-Based Sandbox Approach to Bridge the Privacy Gap among Attitudes, Goals, Knowledge, and Behaviors, Proceedings of the CHI Conference on Human Factors in Computing Systems (2024). DOI: 10.1145/3613904.3642363

Yuwen Lu et al, From Awareness to Action: Exploring End-User Empowerment Interventions for Dark Patterns in UX, Proceedings of the ACM on Human-Computer Interaction (2024). DOI: 10.1145/3637336

Staff Writer

Our in-house science writing team has prepared this content specifically for Lab Horizons

Leave a Reply

Your email address will not be published. Required fields are marked *