# A language model can search for another model’s harmful replies

> Perez and colleagues automate dialogue red teaming by generating test prompts and screening outputs, surfacing tens of thousands of offensive replies from a 280-billion-parameter chatbot.

Canonical: https://aischolar.cc/blog/language-model-red-teams

[← All articles](https://aischolar.cc/blog)

AISCHOLAR / BLOG

# A language model can search for another model’s harmful replies

Perez and colleagues automate dialogue red teaming by generating test prompts and screening outputs, surfacing tens of thousands of offensive replies from a 280-billion-parameter chatbot.

November 14, 2026

Human-written red-team prompts can only cover so many cases. [Perez and colleagues](https://arxiv.org/abs/2202.03286v1) turn prompt discovery itself into an experiment: one language model generates test questions for a target model, and a classifier screens the replies for offensive content.

### Dialogue as a stress test

The authors compare zero-shot and few-shot prompt generation with reinforcement-learning-based approaches. In dialogue tests, their process surfaced tens of thousands of offensive replies from a 280-billion-parameter chatbot. Prompt engineering also let them probe group-targeted language and potential personal-information leakage.

The generator and classifier inherit biases from their training data, so flagged cases need human inspection. The authors particularly caution against treating generated group descriptions as conclusive: some are irrelevant or biased, and others can miss people the evaluation meant to examine. Their recommendation is specific—use model-generated examples to find candidates for human review, especially in group-bias audits.

### Bibliography

-   Perez, E., Huang, S., Song, F., Cai, T., Ring, R., Aslanides, J., Glaese, A., McAleese, N., & Irving, G. (2022). _Red Teaming Language Models with Language Models_. arXiv:2202.03286v1. [https://arxiv.org/pdf/2202.03286](https://arxiv.org/pdf/2202.03286)

Original source: [https://arxiv.org/pdf/2202.03286](https://arxiv.org/pdf/2202.03286)

About AIScholar

## Study LLM behavior.

AIScholar is a research workspace for controlled, replicated studies of LLM behavior using single-turn text responses from OpenRouter-served models. The nine-stage workflow includes response coding, visualization, analysis and write-up; researchers validate measurements and interpret results.

[Explore AIScholar ↗](https://aischolar.cc/features)
