Distinguishing agent traffic from human browsing for pricing

Ascertain whether websites can reliably distinguish traffic generated by autonomous web agents from traffic generated by human users, and, if reliable distinction is not feasible, identify adjustments to the Unbrowse Tier 2 opt-in per-execution pricing model that prevent charging human users in mixed-traffic scenarios.

Background

Tier 2 of Unbrowse’s payment architecture allows site owners to attach per-execution micropayments to API calls hitting their endpoints. This model presumes that agent-generated traffic can be identified and charged without affecting human users.

The authors explicitly flag uncertainty about whether websites can robustly distinguish agent traffic from human traffic and note that the pricing model would need to account for mixed traffic if such distinction is unreliable.

References

A key open question is whether websites can reliably distinguish agent traffic from human browsing; if not, the opt-in pricing model must account for mixed traffic to avoid charging human users.

— Internal APIs Are All You Need: Shadow APIs, Shared Discovery, and the Case Against Browser-First Agent Architectures  (2604.00694 - Tham et al., 1 Apr 2026) in Future Directions, Section "Discussion" (item 3)