# Inherited Goal Drift: Contextual Pressure Can Undermine Agentic Goals

arXiv: https://arxiv.org/abs/2603.03258 · HTML: https://mlsanity.com/paper/2603.03258/

## Key facts

- Cited by 9 papers in the map (data through 14 Oct 2026).
- Placed in General LLM agent frameworks and benchmarks, one of 16 directions in LLM agents and tool use.
- Builds on 9 of its 14 references that are in the map.
- First posted on arXiv 3 Mar 2026.

## Placed in

- LLM agents and tool use / [General LLM agent frameworks and benchmarks](https://mlsanity.com/agents/general-llm-agent-frameworks-and-benchmarks/)

## Builds on (most cited first, 9 of 9)

- [LLM Evaluators Recognize and Favor Their Own Generations](https://mlsanity.com/paper/2404.13076/) (2024)
- [The Instruction Hierarchy: Training LLMs to Prioritize Privileged Instructions](https://mlsanity.com/paper/2404.13208/) (2024)
- [Defending Against Indirect Prompt Injection Attacks With Spotlighting](https://mlsanity.com/paper/2403.14720/) (2024)
- [Large Language Models Often Know When They Are Being Evaluated](https://mlsanity.com/paper/2505.23836/) (2025)
- [Pride and Prejudice: LLM Amplifies Self-Bias in Self-Refinement](https://mlsanity.com/paper/2402.11436/) (2024)
- [Vending-Bench: A Benchmark for Long-Term Coherence of Autonomous Agents](https://mlsanity.com/paper/2502.15840/) (2025)
- [Instructional Segment Embedding: Improving LLM Safety with Instruction Hierarchy](https://mlsanity.com/paper/2410.09102/) (2024)
- [Control Illusion: The Failure of Instruction Hierarchies in Large Language Models](https://mlsanity.com/paper/2502.15851/) (2025)
- [Drift No More? Context Equilibria in Multi-Turn LLM Interactions](https://mlsanity.com/paper/2510.07777/) (2025)

## Built on by (most cited first, 9 of 9)

- [Signal-Driven Observation for Long-Horizon Web Agents](https://mlsanity.com/paper/2606.06708/) (2026)
- [A Taxonomy of Cognitive Capability Gaps in Generative and Agentic AI](https://mlsanity.com/paper/2608.02553/) (2026)
- [The LLM Proposes, the Executive Disposes: A Self-Verifying Agent Instrument that Dissociates Commitment Drift from Binding Drift in Long-Horizon Agents](https://mlsanity.com/paper/2608.04066/) (2026)
- [Same Facts, Different Updates: Inference Setup Shapes LLM Behavior in Medical Allocation](https://mlsanity.com/paper/2608.18108/) (2026)
- [Intent Drift at SME Scale: Deployment Practice, Not Model Capability, Determines Agentic Compliance](https://mlsanity.com/paper/2609.05975/) (2026)
- [When the Agent Becomes the Kernel: A Systematization of Security on the Path to AI-Native Operating Systems](https://mlsanity.com/paper/2609.23700/) (2026)
- [BazaarBench: Delegation Safety in Decentralized C2C Marketplaces Run by LLM Agents](https://mlsanity.com/paper/2610.06748/) (2026)
- [AgentSWE: Can Coding Agents Build the Agent You Actually Want?](https://mlsanity.com/paper/2610.13361/) (2026)
- [New Model, Old Habits: Behavior Transfer under Model Swapping in Multi-Session LLM Agents](https://mlsanity.com/paper/2610.19712/) (2026)

## Abstract (arXiv)

The accelerating adoption of language models (LMs) as agents for deployment in long-context tasks motivates a thorough understanding of goal drift: agents' tendency to deviate from an original objective. While prior-generation language model agents have been shown to be susceptible to drift, the extent to which drift affects more recent models remains unclear. In this work, we provide an updated characterization of the extent and causes of goal drift. We investigate drift in state-of-the-art models within a simulated stock-trading environment (Arike et al., 2025). These models are largely shown to be robust even when subjected to adversarial pressure. We show, however, that this robustness is brittle: across multiple settings, the same models often inherit drift when conditioned on prefilled trajectories from weaker agents. The extent of conditioning-induced drift varies significantly by model family, with only GPT-5.1 maintaining consistent resilience among tested models. We find that drift behavior is inconsistent between prompt variations and correlates poorly with instruction hierarchy following behavior, with strong hierarchy following failing to reliably predict resistance to drift. Finally, we run analogous experiments in a new emergency room triage environment to show preliminary evidence for the transferability of our results across qualitatively different settings. Our findings underscore the continued vulnerability of modern LM agents to contextual pressures and the need for refined post-training techniques to mitigate this.