Observation · Perplexity

How do AI search engines choose which sources to cite?

Perplexity described source selection as passage-level retrieval and re-ranking, followed by answer generation with attribution. It emphasized question relevance, extractability, credibility, corroboration and freshness.

15 November 2026, 17:10 CSTen-US (English answer; zh-CN interface)Run 1 of 3Method v0.1
Question

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, 17:10 CST
Fresh conversation
Yes
Answer snapshot

What the answer said

Perplexity answered that AI search engines generally cite sources whose specific passages best answer the question, while weighing credibility, freshness and traceability. It described a four-stage process: understand the query, retrieve candidate pages, re-rank passages, then generate an answer with citations. It listed direct relevance, an extractable answer, specific evidence, trust, corroboration and freshness as factors. It also cautioned that a citation shows grounding in a retrieved source; it does not prove that every sentence was independently verified.

Linked citations

10 recorded sources

Limitations

What this observation cannot show

One authenticated free-account run on 15 November 2026. Perplexity did not expose its automatic model version or retrieval index. The answer can change by session, language, account state, model selection and date.