# Great Models Think Alike and this Undermines AI Oversight

arXiv: https://arxiv.org/abs/2502.04313 · HTML: https://mlsanity.com/paper/2502.04313/

## Key facts

- Cited by 58 papers in the map (data through 14 Oct 2026).
- Placed in Reasoning benchmarks and evaluation, one of 17 directions in LLM reasoning.
- Builds on 0 of its 106 references that are in the map.
- First posted on arXiv 6 Feb 2025.

## Placed in

- LLM reasoning / [Reasoning benchmarks and evaluation](https://mlsanity.com/lr/reasoning-benchmarks-and-evaluation/)

## Built on by (most cited first, 30 of 58)

- [Correlated Errors in Large Language Models](https://mlsanity.com/paper/2506.07962/) (2025)
- [Answer Matching Outperforms Multiple Choice for Language Model Evaluation](https://mlsanity.com/paper/2507.02856/) (2025)
- [Do LLM Evaluators Prefer Themselves for a Reason?](https://mlsanity.com/paper/2504.03846/) (2025)
- [How Benchmark Prediction from Fewer Data Misses the Mark](https://mlsanity.com/paper/2506.07673/) (2025)
- [When Your AI Agent Succumbs to Peer-Pressure: Studying Opinion-Change Dynamics of LLMs](https://mlsanity.com/paper/2510.19107/) (2025)
- [Stop Automating Peer Review Without Rigorous Evaluation](https://mlsanity.com/paper/2605.03202/) (2026)
- [Discrepancies are Virtue: Weak-to-Strong Generalization through Lens of Intrinsic Dimension](https://mlsanity.com/paper/2502.05075/) (2025)
- [PrivacyBench: A Conversational Benchmark for Evaluating Privacy in Personalized AI](https://mlsanity.com/paper/2512.24848/) (2025)
- [PiCSAR: Probabilistic Confidence Selection And Ranking for Reasoning Chains](https://mlsanity.com/paper/2508.21787/) (2025)
- [The Hot Mess of AI: How Does Misalignment Scale With Model Intelligence and Task Complexity?](https://mlsanity.com/paper/2601.23045/) (2026)
- [Consensus is Not Verification: Why Crowd Wisdom Strategies Fail for LLM Truthfulness](https://mlsanity.com/paper/2603.06612/) (2026)
- [Are Diversity Metrics Measuring Diversity? A Capability-Controlled Audit of Majority-Vote Gain in LLM Ensembles](https://mlsanity.com/paper/2607.20768/) (2026)
- [Quantitative LLM Judges](https://mlsanity.com/paper/2506.02945/) (2025)
- [Mitigating Self-Preference by Authorship Obfuscation](https://mlsanity.com/paper/2512.05379/) (2025)
- [Seven simple steps for log analysis in AI systems](https://mlsanity.com/paper/2604.09563/) (2026)
- [When More Sampling Hurts: The Modal Ceiling and Correlation Ceiling of Test-Time Scaling](https://mlsanity.com/paper/2606.28661/) (2026)
- [Decision Models vs. LLMs as Rubric Judges: Cheaper, Faster, and Wrong in the Same Places](https://mlsanity.com/paper/2609.29769/) (2026)
- [Decentralized Arena: Towards Democratic and Scalable Automatic Evaluation of Language Models](https://mlsanity.com/paper/2505.12808/) (2025)
- [Weak-to-Strong Generalization via Bregman Bias-Variance Decomposition](https://mlsanity.com/paper/2505.24313/) (2025)
- [Contrastive Decoding Mitigates Score Range Bias in LLM-as-a-Judge](https://mlsanity.com/paper/2510.18196/) (2025)
- [CIAware-Bench: Benchmarking Control Intervention Awareness Across Frontier LLMs](https://mlsanity.com/paper/2606.11063/) (2026)
- [Does Weak-to-strong Generalization Happen under Spurious Correlations?](https://mlsanity.com/paper/2509.24005/) (2025)
- [Dependence-Aware Label Aggregation for LLM-as-a-Judge via Ising Models](https://mlsanity.com/paper/2601.22336/) (2026)
- [The Subjectivity of Monoculture](https://mlsanity.com/paper/2602.24086/) (2026)
- [Knowledge Divergence and the Value of Debate for Scalable Oversight](https://mlsanity.com/paper/2603.05293/) (2026)
- [Intersectional Sycophancy: How Perceived User Demographics Shape False Validation in Large Language Models](https://mlsanity.com/paper/2604.11609/) (2026)
- [Cutting LLM Evaluation Costs with SySRs: A Bandit Algorithm that Provably Exploits Model Similarity](https://mlsanity.com/paper/2606.07726/) (2026)
- [State-dependent error correlations shape voting thresholds in committees of AI agents](https://mlsanity.com/paper/2607.23931/) (2026)
- [Language Models Agree With Each Other, Not With Readers](https://mlsanity.com/paper/2607.29274/) (2026)
- [What if I ask in textitalia lingua? Measuring Functional Similarity Across Languages](https://mlsanity.com/paper/2509.04032/) (2025)

## Abstract (arXiv)

As Language Model (LM) capabilities advance, evaluating and supervising them at scale is getting harder for humans. There is hope that other language models can automate both these tasks, which we refer to as ''AI Oversight''. We study how model similarity affects both aspects of AI oversight by proposing Chance Adjusted Probabilistic Agreement (CAPA): a metric for LM similarity based on overlap in model mistakes. Using CAPA, we first show that LLM-as-a-judge scores favor models similar to the judge, generalizing recent self-preference results. Then, we study training on LM annotations, and find complementary knowledge between the weak supervisor and strong student model plays a crucial role in gains from ''weak-to-strong generalization''. As model capabilities increase, it becomes harder to find their mistakes, and we might defer more to AI oversight. However, we observe a concerning trend -- model mistakes are becoming more similar with increasing capabilities, pointing to risks from correlated failures. Our work underscores the importance of reporting and correcting for model similarity, especially in the emerging paradigm of AI oversight.