Real Safety AI Foundation / Research

Privacy, Surveillance, and Consent

Never a User: The Third Party the Consent Apparatus Cannot See

Publisher: Real Safety AI Foundation

Working paper. Not peer reviewed.

Written
August 2026
Pages
10
Download PDFPlain language slides (PDF)Export BibTeXExport RIS

Abstract

Contemporary governance of consumer artificial intelligence is keyed almost entirely to the user: the person who opened the application, accepted the terms, and can therefore be said to have agreed. That keying is not a drafting habit. It is the load-bearing assumption of the whole apparatus, and it produces a structural blind spot rather than an occasional gap. The person standing in the room while a wearable device captures him, the person on the other end of a conversation an assistant is coaching, and the person killed by someone else's interaction with a chatbot share one property: they never agreed to anything, and every instrument built to protect people from these systems begins by asking whether they did. This paper names that population and argues that its invisibility is architectural. It proceeds in four steps. First, it shows that American products liability abandoned the user limitation more than fifty years ago and did so deliberately, across at least six jurisdictions in under a decade, so the doctrinal resources for reaching the non-user are neither novel nor contested. Second, it shows that the consent architecture governing consumer artificial intelligence has no corresponding move available, because consent is the mechanism by which the apparatus identifies whom it protects. Third, it demonstrates that the obvious repair, a bystander opt-out, is structurally impossible in the ambient-capture case rather than merely burdensome, since determining whether a person has opted out requires the identifying operation the opt-out exists to prevent. Fourth, it argues that the resulting exposure is not distributed evenly, and that the people least able to remove themselves from an instrumented environment absorb it. The contribution is not a proposed remedy. It is a diagnosis: the field has adopted a proxy, agreement, for a thing that matters, exposure, and has been enforcing the proxy against the people for whom it never held.

Keywords

  • bystander liability
  • non-user harm
  • consent architecture
  • ambient computing
  • third-party exposure
  • strict products liability
  • opt-out infeasibility
  • wearable capture
  • technology governance
  • disability and access
  • human-computer interaction
  • informed agreement

Plain language slides

First slide of the plain language summary of Never a User: The Third Party the Consent Apparatus Cannot SeeOpen the 20-slide summary (PDF)

Suggested citation

Gilly, Travis. "Never a User: The Third Party the Consent Apparatus Cannot See." Real Safety AI Foundation Working Paper, August 2026. https://realsafetyai.org/research/never-a-user/

Other versions

This paper is also posted on SSRN.

SSRN version

References (25)

  1. Aaraj, J. A., et al. (2025). BLINDSPOT: Enabling bystander-controlled privacy signaling for camera-enabled devices [Preprint].
  2. Ahmad, I., Farzan, R., Kapadia, A., & Lee, A. J. (2020). Tangible privacy: Towards user-centric sensor designs for bystander privacy. Proceedings of the ACM on Human-Computer Interaction, 4(CSCW2), 1–28.
  3. American Law Institute. (1965). Restatement (Second) of Torts § 402A.
  4. Baruh, L., & Popescu, M. (2017). Big data analytics and the limits of privacy self-management. New Media & Society, 19(4), 579–596.
  5. Codling v. Paglia, 32 N.Y.2d 330, 298 N.E.2d 622 (1973).
  6. Elmore v. American Motors Corp., 70 Cal. 2d 578, 451 P.2d 84 (1969).
  7. Garcia v. Character Technologies, Inc., 785 F. Supp. 3d 1157 (M.D. Fla. 2025).
  8. Giberson v. Ford Motor Co., 504 S.W.2d 8 (Mo. 1974).
  9. Howes v. Hansen, 56 Wis. 2d 247, 201 N.W.2d 825 (1972).
  10. Hoyle, R., Templeman, R., Armes, S., Anthony, D., Crandall, D., & Kapadia, A. (2014). Privacy behaviors of lifeloggers using wearable cameras. In Proceedings of the 2014 ACM International Joint Conference on Pervasive and Ubiquitous Computing (pp. 571–582). ACM.
  11. Hull, G. (2015). Successful failure: What Foucault can teach us about privacy self-management in a world of Facebook and big data. Ethics and Information Technology, 17(2), 89–101.
  12. Khawaja, Y., et al. (2026). See me if you can: A multi-layer protocol for bystander privacy with consent-based restoration. In Proceedings of the 2026 CHI Conference on Human Factors in Computing Systems. ACM.
  13. Koelle, M., Wolf, K., & Boll, S. (2018). Your smart glasses’ camera bothers me! Exploring opt-in and opt-out gestures for privacy mediation. In Proceedings of the Tenth Nordic Conference on Human-Computer Interaction (pp. 473–481). ACM.
  14. Kröger, J. L., Lutz, O. H.-M., & Ullrich, S. (2021). The myth of individual control: Mapping the limitations of privacy self-management [Unpublished manuscript].
  15. Lee, K., Hong, J., Jarjue, E., Mensah, E. A., & Kacorri, H. (2020). Pedestrian detection with wearable cameras for the blind: A two-way perspective. In Proceedings of the 2020 CHI Conference on Human Factors in Computing Systems (pp. 1–12). ACM.
  16. Mitchell v. Miller, 26 Conn. Super. Ct. 142, 214 A.2d 694 (1965).
  17. O’Hagan, J., Saeghe, P., Gugenheimer, J., Medeiros, D., Marky, K., Khamis, M., & McGill, M. (2022). Privacy-enhancing technology and everyday augmented reality. Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies, 6(4), 1–35.
  18. Passwaters v. General Motors Corp., 454 F.2d 1270 (8th Cir. 1972).
  19. Pierce, J., Fox, S., Merrill, N., & Wong, R. (2022). Addressing adjacent actor privacy: Designing for bystanders, co-users, and surveilled subjects of smart home cameras. In Proceedings of the 2022 ACM Designing Interactive Systems Conference (pp. 26–40). ACM.
  20. Saqib, E., Aljasim, S., Zhao, V. Y., & Das, S. (2025). Bystander privacy in smart homes: A systematic review of concerns and solutions. ACM Transactions on Computer-Human Interaction, 32(1), 1–33.
  21. Schaub, F., Balebako, R., & Cranor, L. F. (2017). Designing effective privacy notices and controls. IEEE Internet Computing, 21(3), 70–77.
  22. Sloan, R. H., & Warner, R. (2013). Beyond notice and choice: Privacy, norms, and consent. Journal of High Technology Law, 14, 370–412.
  23. Solove, D. J. (2013). Introduction: Privacy self-management and the consent dilemma. Harvard Law Review, 126, 1880–1903.
  24. Sprigman, C. J., Marotta-Wurgler, F., & Levine, D. (2024). Privacy decision-making and the effects of privacy choice architecture: Experiments toward the design of behaviorally-aware privacy regulation. Journal of Empirical Legal Studies, 21(4), 690–735.
  25. Vosburg v. Putney, 80 Wis. 523, 50 N.W. 403 (1891).

All research