Recorded prompt
“What causes inaccurate AI citations?”
- Product
- Perplexity
- Surface
- Perplexity Search with web search enabled
- Model
- Perplexity automatic model (model version not exposed)
- Locale
- en-US (English answer; zh-CN interface)
- Observed answer
- Open session ↗
- Observed
- 15 November 2026, 21:06 CST
- Fresh conversation
- Yes
What the answer said
10秒 AI citations go wrong mainly because language models generate text probabilistically: they can produce plausible-looking sources without verifying that the work exists or that its details are correct. Errors are more likely when training data is noisy, source information is incomplete, retrieval finds the wrong material, or the system lacks a fact-checking step. acm +1
10 recorded sources
What this observation cannot show
Authenticated free-account Perplexity browser run in a fresh conversation. The product did not expose its complete retrieval configuration. The source panel reported 10 sources and 10 unique source records were captured. Results can vary by session, locale, account state and date.