Paper maps 84 suspected cases of agent-driven demand surges on government services across 11 jurisdictions

Sources: primary, discussion
Summary: The paper catalogues 84 suspected incidents across 11 jurisdictions in which automated agents acting for users drove demand surges on public services such as appointment booking and application intake. The authors map possible government responses to that flooding and identify friction-inducing measures such as fees as the fastest to deploy. The arXiv record states the paper is to appear in the proceedings of the 9th AAAI Conference on AI, Ethics, and Society, held 2026-10-12 to 2026-10-14.
Why it matters: The authors argue the fastest government response, friction such as fees, trades away equitable access, which is the constraint on anyone building an agent that files applications on a user's behalf.
Follow-up: Whether the incident set is confirmed by the named agencies.