Who controls AI conversations?

Every interaction with generative AI begins with an invisible layer: system prompts. They shape how AI systems respond, what they prioritize, and ultimately how users experience these technologies – often without users even being aware of their influence.

This is exactly where the award-winning research of Anna Neumann begins. Even before the official conference in April, her paper Who Controls the Conversation? User Perspectives on Generative AI (LLM) System Prompts has been recognized with a Best Paper Award at the ACM CHI Conference on Human Factors in Computing Systems 2026–one of the most prestigious venues in human-computer interaction.

Anna Neumann is a PhD researcher in the Compliant and Accountable Systems group at RC Trust, supervised by Prof. Jat Singh. The work was conducted in close collaboration with Yulu Pi and Jat Singh, reflecting the group’s interdisciplinary approach at the intersection of technology, society, and governance.

The paper addresses a fundamental question for the future of AI: how much control and transparency should users have over the systems they interact with?

To answer this, the research combines qualitative, quantitative, and computational methods. By analyzing 1,309 real-world system prompts and complementing this with user studies, the team provides one of the most comprehensive investigations into how these hidden instructions operate–and how they are perceived.

The work makes three key contributions.

First, it introduces a taxonomy of seven types of system prompts, offering a structured way to understand how these systems are configured in practice.

Second, it provides empirical insights into user perspectives. The findings show a clear demand for greater agency: 89% of participants want transparency about system prompts, and 79% want some form of control. At the same time, values such as privacy and freedom from bias emerge as central concerns.

Third, the paper translates these findings into concrete design and governance recommendations. It outlines how meaningful transparency mechanisms, participatory design processes, and governance structures–such as standards, provenance, and evaluation–could enable more inclusive and accountable AI systems.

Taken together, the research opens up a new direction for user-centered system prompt design. While system prompts already shape the everyday interactions of millions of users, they remain largely invisible and difficult to contest. Making them more transparent and accountable is therefore not just a technical challenge, but a societal one.

With this award, RC Trust highlights the importance of interdisciplinary research that connects technical innovation with questions of responsibility, governance, and human agency in AI.

Anna Neumann will present the paper at CHI 2026 on Wednesday, 15 April, from 9:12 to 9:24 AM.