# Reliability Check: An Analysis of GPT-3's Response to Sensitive Topics and Prompt Wording

arXiv: https://arxiv.org/abs/2306.06199 · HTML: https://mlsanity.com/paper/2306.06199/

## Key facts

- Cited by 13 papers in the map (data through 14 Oct 2026).
- Builds on 17 of its 39 references that are in the map.
- First posted on arXiv 9 Jun 2023.

## Builds on (most cited first, 17 of 17)

- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [TruthfulQA: Measuring How Models Mimic Human Falsehoods](https://mlsanity.com/paper/2109.07958/) (2021)
- [Learning to summarize from human feedback](https://mlsanity.com/paper/2009.01325/) (2020)
- [The Pile: An 800GB Dataset of Diverse Text for Language Modeling](https://mlsanity.com/paper/2101.00027/) (2020)
- [Evaluating the Factual Consistency of Abstractive Text Summarization](https://mlsanity.com/paper/1910.12840/) (2019)
- [KILT: a Benchmark for Knowledge Intensive Language Tasks](https://mlsanity.com/paper/2009.02252/) (2020)
- [Measuring and Improving Consistency in Pretrained Language Models](https://mlsanity.com/paper/2102.01017/) (2021)
- ["Liar, Liar Pants on Fire": A New Benchmark Dataset for Fake News Detection](https://mlsanity.com/paper/1705.00648/) (2017)
- [Prompting GPT-3 To Be Reliable](https://mlsanity.com/paper/2210.09150/) (2022)
- [Entity-level Factual Consistency of Abstractive Text Summarization](https://mlsanity.com/paper/2102.09130/) (2021)
- [Assessing The Factual Accuracy of Generated Text](https://mlsanity.com/paper/1905.13322/) (2019)
- [Evaluating the Factual Consistency of Large Language Models Through News Summarization](https://mlsanity.com/paper/2211.08412/) (2022)
- [Is GPT-3 Text Indistinguishable from Human Text? Scarecrow: A Framework for Scrutinizing Machine Text](https://mlsanity.com/paper/2107.01294/) (2021)
- [A Richly Annotated Corpus for Different Tasks in Automated Fact-Checking](https://mlsanity.com/paper/1911.01214/) (2019)
- [Improving Factual Consistency of Abstractive Summarization via Question Answering](https://mlsanity.com/paper/2105.04623/) (2021)
- [CheckThat! at CLEF 2020: Enabling the Automatic Identification and Verification of Claims in Social Media](https://mlsanity.com/paper/2001.08546/) (2020)

## Built on by (most cited first, 13 of 13)

- [Securing Large Language Models: Threats, Vulnerabilities and Responsible Practices](https://mlsanity.com/paper/2403.12503/) (2024)
- [Disinformation Capabilities of Large Language Models](https://mlsanity.com/paper/2311.08838/) (2023)
- [Don't Go To Extremes: Revealing the Excessive Sensitivity and Calibration Limitations of LLMs in Implicit Hate Speech Detection](https://mlsanity.com/paper/2402.11406/) (2024)
- [A Study on Large Language Models' Limitations in Multiple-Choice Question Answering](https://mlsanity.com/paper/2401.07955/) (2024)
- [Decoding the AI Pen: Techniques and Challenges in Detecting AI-Generated Text](https://mlsanity.com/paper/2403.05750/) (2024)
- [TruthEval: A Dataset to Evaluate LLM Truthfulness and Reliability](https://mlsanity.com/paper/2406.01855/) (2024)
- [On Meta-Prompting](https://mlsanity.com/paper/2312.06562/) (2023)
- [Detection of Conspiracy Theories Beyond Keyword Bias in German-Language Telegram Using Large Language Models](https://mlsanity.com/paper/2404.17985/) (2024)
- [Structuring the Space of Sociotechnical Alignment](https://mlsanity.com/paper/2607.01250/) (2026)
- [Exploring Robustness of LLMs to Paraphrasing Based on Sociodemographic Factors](https://mlsanity.com/paper/2501.08276/) (2025)
- [LLM Echo Chamber: personalized and automated disinformation](https://mlsanity.com/paper/2409.16241/) (2024)
- [Metric assessment protocol in the context of answer fluctuation on MCQ tasks](https://mlsanity.com/paper/2507.15581/) (2025)
- [From Interpretability to Control: Insights from Six Years of the TrustNLP Workshop](https://mlsanity.com/paper/2608.11171/) (2026)

## Abstract (arXiv)

Large language models (LLMs) have become mainstream technology with their versatile use cases and impressive performance. Despite the countless out-of-the-box applications, LLMs are still not reliable. A lot of work is being done to improve the factual accuracy, consistency, and ethical standards of these models through fine-tuning, prompting, and Reinforcement Learning with Human Feedback (RLHF), but no systematic analysis of the responses of these models to different categories of statements, or on their potential vulnerabilities to simple prompting changes is available. In this work, we analyze what confuses GPT-3: how the model responds to certain sensitive topics and what effects the prompt wording has on the model response. We find that GPT-3 correctly disagrees with obvious Conspiracies and Stereotypes but makes mistakes with common Misconceptions and Controversies. The model responses are inconsistent across prompts and settings, highlighting GPT-3's unreliability. Dataset and code of our analysis is available in https://github.com/tanny411/GPT3-Reliability-Check.