# The Bureaucracy of Speed: Structural Equivalence Between Memory Consistency Models and Multi-Agent Authorization Revocation

arXiv: https://arxiv.org/abs/2603.09875 · HTML: https://mlsanity.com/paper/2603.09875/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Placed in Agent threats, guardrails and governance, one of 16 directions in LLM agents and tool use.
- First posted on arXiv 10 Mar 2026.

## Placed in

- LLM agents and tool use / [Agent threats, guardrails and governance](https://mlsanity.com/agents/agent-threats-guardrails-and-governance/)

## Built on by (most cited first, 1 of 1)

- [Fault-Tolerant Budget Conservation in Distributed Multi-Agent Delegation](https://mlsanity.com/paper/2610.00349/) (2026)

## Abstract (arXiv)

The temporal assumptions underpinning conventional Identity and Access Management collapse under agentic execution regimes. A sixty-second revocation window permits on the order of 6 × 103 unauthorized API calls at 100 ops/tick; at AWS Lambda scale, the figure approaches 6 × 105. This is a coherence problem, not merely a latency problem. We define a Capability Coherence System (CCS) and construct a state-mapping varphi : Sigmarm MESI → Sigmarm auth preserving transition structure under bounded-staleness semantics. A safety theorem bounds unauthorized operations for the execution-count Release Consistency-directed Coherence (RCC) strategy at Drm rcc ≤ n, independent of agent velocity v -- a qualitative departure from the O(v · TTL) scaling of time-bounded strategies. Tick-based discrete event simulation across three business-contextualised scenarios (four strategies, ten deterministic seeds each) confirms: RCC achieves a 120× reduction versus TTL-based lease in the high-velocity scenario (50 vs. 6,000 unauthorized operations), and 184× under anomaly-triggered revocation. Zero bound violations across all 120 runs confirm the per-capability safety guarantee. Simulation code: https://github.com/hipvlady/prizm