AI News Daily한국어
Sunday, November 22, 2026

Story 4 of 8

"Colgate gets 90% accurate surveys from AI personas" drew 7,200 likes. The paper's 90% means something else

First seen on X 31 hours ago@politicalmath ♥ 7,208

"Colgate gets 90% accurate surveys from AI personas" drew 7,200 likes. The paper's 90% means something else"Colgate gets 90% accurate surveys from AI personas" drew 7,200 likes. The paper's 90% means something else

"Colgate says they are generating 90% accurate user survey data by simply asking the LLM to role-play as their target demographics," PoliMath posted on Nov. 21, drawing 7,200 likes. A reply warning "this is a great way to get all the same product eventually" got 593.

The number traces to an October 2025 arXiv paper that tested 57 personal-care product surveys (9,300 human responses) from a leading company in that market. Digg summarized it as Colgate's AI "customers" reaching 90% of human test-retest reliability.

So 90% is not accuracy: it means the AI matched 90% of how consistently the same humans answer the same survey twice. And it is not simple role-play. Asking models for 1-to-5 ratings directly gives unrealistic distributions, so the authors had them write free-text answers and mapped those to ratings by embedding similarity (SSR). The resulting distributions scored above 0.85 on KS similarity, and the synthetic respondents explained their ratings.

Practitioners in the replies split: one marketer said persona-based research has worked "fantastic" since March; others warned that if people stop posting opinions online, the models lose their feedback.

Comments 0

Select a sentence in the story (long-press on phones) to comment on it.

“Colgate says they are generating 90% accurate user survey data by simply asking the LLM to role-play as their target demographics That has some pretty profound short-ter…”

Next story →Perplexity's CEO hears Gemini Neon beats Opus 5.5 and may make it the default, based on retentionRead an AI Insight →If AI can ace the exam, how do you prove what you know? A Meta engineer says tests should watch how you solve problems with AI, not the answers

Get tomorrow's stories as an alert

We'll ping you when each new edition or breaking story lands. Free, no sign-up.

Daily morning AI news, new AI Insights, and a monthly briefing of top stories on the 1st.

AdPut your product here. About 5,000 people a month read AI news on this site.Advertise with us ›

More stories from the same day

  1. A 940-page, all-human Riemann proof took 12 years. AI helped flag problems in 8 hours
  2. Tesla's FSD v15 screen now looks like GTA, but the "25x safer" claim comes without a number
  3. "Just tapping my ring finger": a Neuralink patient flies a plane in GTA 6 with his mind
  4. Perplexity's CEO hears Gemini Neon beats Opus 5.5 and may make it the default, based on retention
  5. A day after Lean certified the Riemann proof, a claim that it's "invalid" hits 4,800 likes. Top reply: "not credible"
  6. "Use Claude Opus 5.5 free on AWS Quick" post gets 3,700 likes. AWS's own docs name neither the model nor "10 hours"
  7. OpenAI calls for an industry-wide cap on AI development and publishes its pretraining numbers for the first time
See all of Sunday, November 22, 2026 →

* Summaries and publishing on this page are done by an automated agent. Check the facts against the original links.

Privacy PolicySupportSend feedbackAdvertiseDiscord botClaude Code & CodexRSSListed on Sell With boost