Privacy Preference Controls for Agentic Systems

arXiv:2610.29929v1 Announce Type: new Abstract: In response to vast and widespread online tracking, regulators and technologists alike have worked to develop laws and tools that allow people to express their privacy preferences. These measures seek to offer users greater control over the sale and sharing of personal information, putting agency in their hands to protect their privacy. However, the web is undergoing a seismic shift, away from strictly \textit{user-driven} behavior towards \textit{agentic behavior}, where AI agents take autonomous actions on behalf of users. In this study, we ask: What should privacy preference controls look like in the agentic web and agentic systems more generally? To answer this question, we develop a taxonomy of enforceable privacy preferences for agentic systems. Our taxonomy is informed both by laws and empirical studies of existing privacy tools as well as an audit of 30 leading AI assistants and agent platforms. To explore how our taxonomy can be implemented, we design and develop eight open-source prototypes that mirror real-world agentic applications and demonstrate how major aspects of our taxonomy can be structurally integrated in agentic protocols, such as the Model Control Protocol and Agent2Agent protocol. Ultimately, our study seeks to provide the groundwork for ensuring end-user privacy and autonomy is not lost in the newly arriving agentic world.

Study highlights need for privacy controls in agentic systems.

The study addresses a critical emerging issue: protecting user privacy and autonomy as AI agents act independently on users' behalf. By developing privacy control taxonomies and prototypes, it offers mechanisms for protection. However, the shift to agentic systems also raises concerns about loss of control and potential misuse, balancing positive and negative implications.

This is constructive research addressing a real emerging gap, privacy control in agentic AI, with concrete prototypes and protocol integration. It identifies a risk (loss of user autonomy) but also offers a mitigating tool. Net implication is mildly positive, contingent on adoption, which is uncertain.

This academic paper identifies a significant upcoming challenge concerning user privacy in an AI-driven 'agentic' web. It proposes a taxonomy and open-source prototypes to address these concerns proactively. While it highlights a potential future risk, the study's focus on developing solutions positions it optimistically on the spectrum by offering a path to mitigate that risk.

The paper proposes a structured approach to maintaining user privacy in AI-driven systems. It offers concrete steps and prototypes to achieve this, suggesting a positive impact on user autonomy and privacy.

SOURCE     arXiv cs.CY (Computers and Society) (tier 1)
   ↓
DOCUMENT   2a5baa77-27d9-4e4d-a140-8121f57b680f
           https://arxiv.org/abs/2610.29929
   ↓
EVIDENCE   2 extracted excerpts
   ↓
MODEL RUN  5 runs, methodology 1.0.1
   ↓
SCORE      +12  (Mixed or uncertain)
   ↓
CONFIDENCE 78%