Why Agents Compromise Safety Under Pressure

Abstract

Large Language Model agents deployed in complex environments frequently encounter a conflict between maximizing goal achievement and adhering to safety constraints. This paper identifies a new concept called Agentic Pressure, which characterizes the endogenous tension emerging when compliant execution becomes infeasible. We demonstrate that under this pressure agents exhibit normative drift where they strategically sacrifice safety to preserve utility. Notably we find that advanced reasoning capabilities accelerate this decline as models construct linguistic rationalizations to justify violation. Finally, we analyze the root causes and explore preliminary mitigation strategies, such as pressure isolation, which attempts to restore alignment by decoupling decision-making from pressure signals.

Figures (6)

• Figure 1: The "Good Agent" Paradox: While the user's request is non-malicious, the combination of high urgency and resource deadlock forces the agent to trade off safety for goal achievement.
• Figure 2: Taxonomy of Pressure Sources
• Figure 3: Preliminary results on TravelPlanner under non-adversarial pressure
• Figure 4: Overview of the Agentic Pressure Evaluation Framework.
• Figure 5: Normative Drift Distribution. The scatter plot shows individual episode outcomes, highlighting the primary shift from the Ideal Region (Safety, Utility) to the Drift Region (Low Safety, Higher Utility) under agentic pressure.