[2302.03494] A Categorical Archive of ChatGPT Failures

Large language models have been demonstrated to be valuable in different fields. ChatGPT, developed by OpenAI, has been trained using massive amounts of data and simulates human conversation by comprehending context and generating appropriate responses. It has garnered significant attention due to its ability to effectively answer a broad range of human inquiries, with fluent and comprehensive answers surpassing prior public chatbots in both security and usefulness. However, a comprehensive analysis of ChatGPT's failures is lacking, which is the focus of this study. Ten categories of failures, including reasoning, factual errors, math, coding, and bias, are presented and discussed. The risks, limitations, and societal implications of ChatGPT are also highlighted. The goal of this study is to assist researchers and developers in enhancing future language models and chatbots.

A Categorical Archive of ChatGPT Failures Ali Borji Quintic AI aliborji@gmail.com April 5, 2023 arXiv:2302.03494v8 [cs.CL] 3 Apr 2023…

saved by

• Surabhi Shetty

related reading

• [2201.11903] Chain of Thought Prompting Elicits Reasoning in Large Language Modelsarxiv.org
• [2602.05910] Chunky Post-Training: Data Driven Failures of Generalizationarxiv.org
• ChatGPT Is Dumber Than You Think - The Atlantictheatlantic.com
• [2203.02155] Training language models to follow instructions with human feedbackarxiv.org
• 2405.01470arxiv.org
• Extracting Training Data from ChatGPTnot-just-memorization.github.io
• Collections: On ChatGPT – A Collection of Unmitigated Pedantryacoup.blog
• Language Models can Solve Computer Tasksarxiv.org
• 2308.03958arxiv.org
• The Yale Review | Melanie Mitchell: The Dangerous Unknowns at the…yalereview.org
• 2305.18654arxiv.org
• [2306.13000] Apolitical Intelligence? Auditing Delphi's responses on controversial political issues in the USarxiv.org