ChatGPT不信任闪电队球迷：语境中的护栏敏感性
ChatGPT Doesn't Trust Chargers Fans: Guardrail Sensitivity in Context
AI总结：
本文研究用户语境信息对LLM护栏敏感性的影响，发现GPT-3.5对年轻、女性、亚裔美国人等角色及体育迷身份存在护栏偏见，且会基于用户意识形态推断调整护栏行为。
摘要：
虽然生成式语言模型在生产环境中的偏见已被广泛记录，但其护栏机制的偏见却被忽视。本文研究用户的语境信息如何影响大语言模型（LLM）拒绝执行请求的可能性。通过生成包含意识形态和人口统计信息的用户传记，我们发现GPT-3.5的护栏敏感性存在多种偏见：在请求受审查或非法信息时，年轻、女性和亚裔美国人角色更可能触发拒绝护栏；护栏还具有谄媚性，会拒绝符合用户可能反对的政治立场的请求；某些身份群体和看似无害的信息（如体育迷身份）引发的护栏敏感性变化，与直接的政治意识形态陈述相似；对于每个人口统计类别甚至美式橄榄球队球迷身份，ChatGPT似乎会推断可能的政治意识形态并相应调整护栏行为。
英文摘要：
While the biases of language models in production are extensively documented, the biases of their guardrails have been neglected. This paper studies how contextual information about the user influences the likelihood of an LLM to refuse to execute a request. By generating user biographies that offer ideological and demographic information, we find a number of biases in guardrail sensitivity on GPT-3.5. Younger, female, and Asian-American personas are more likely to trigger a refusal guardrail when requesting censored or illegal information. Guardrails are also sycophantic, refusing to comply with requests for a political position the user is likely to disagree with. We find that certain identity groups and seemingly innocuous information, e.g., sports fandom, can elicit changes in guardrail sensitivity similar to direct statements of political ideology. For each demographic category and even for American football team fandom, we find that ChatGPT appears to infer a likely political ideology and modify guardrail behavior accordingly.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
