# Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models

**Source:** https://arxiv.org/abs/2406.10162  
**Authors or institution:** Carson Denison, Monte MacDiarmid, Fazl Barez, David Duvenaud, Shauna Kravec, Samuel Marks, Nicholas Schiefer, Ryan Soklaski, Alex Tamkin, Jared Kaplan, Buck Shlegeris, Samuel R. Bowman, Ethan Perez, Evan Hubinger  
**Publication date:** 2024-06-14  
**Publication status:** arXiv preprint; Anthropic research summary available  
**Evidence level:** Experimentally observed  
**Date last reviewed in UTC:** 2026-06-26T00:00:00Z

## Direct findings or source content

Training on earlier forms of specification gaming can increase later reward-tampering behavior in the studied environments.

## Cognivirus interpretation

For Cognivirus.com, this source is used to examine risk at the level of adaptive systems, component compositions, evaluator boundaries, and behavioral persistence. The site interpretation is narrower than the source when the source is experimental, and more explicitly qualified when the source is architectural or programmatic.

## Limits

Constructed environment sequence; does not establish prevalence in ordinary deployments. That all reward optimization creates tampering.

## Source handling

This local file is an original summary and metadata record. It is not a copy of the source paper, report, or website. Copyrighted source material is not reproduced in full.