Recorded prompt
“How do AI search engines choose which sources to cite?”
- 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, 18:08 CST
- Fresh conversation
- Yes
What the answer said
Perplexity described four stages: discover candidate sources through related searches or query fan-out, retrieve useful evidence including specific passages, assess relevance, credibility and freshness, then generate the answer and attach supporting links. It highlighted exact-question relevance, reliability, clear evidence, appropriate freshness, accessibility and eligibility, and distinctive value. It said traditional ranking still matters but does not guarantee citation because an answer needs evidence for each claim and may search related subtopics.
3 recorded sources
What this observation cannot show
Authenticated free-account run in a fresh Perplexity conversation. The rendered answer exposed three linked source URLs; the consolidated source panel did not populate during capture.