"Consensus Review 2026: Ask a Question, Get the Scientific Consensus"
Consensus is a search engine that answers questions with the body of scientific literature rather than a single snippet. Ask ‘does intermittent fasting improve insulin sensitivity’ and it returns a consensus rating drawn from many papers, with study cards.
How it works#
You ask a natural-language question; Consensus retrieves studies, uses LLMs to extract each study’s answer and stance, then aggregates a consensus meter (e.g. ‘Strong consensus: Yes’). Each result card shows the claim, study type and a link. The 2026 Copilot mode writes a cited summary paragraph.
Strengths#
- Aggregates stance across many studies
- Consensus meter makes evidence legible
- Great for health, nutrition and social-science claims
- Citations to real papers
- Fast reality-checks of hype
Weaknesses#
- Aggregation can hide disagreement nuance
- Smaller corpus than general search
- Best on yes/no style questions
- Some depth behind paid tier
Pricing#
-
Free — $0: Basic queries
-
Pro — $9-12/mo: More queries, Copilot, export
-
Teams — custom: Shared workspaces
Verdict#
Consensus is the best tool for cutting through health and science hype with cited evidence. See also Elicit for deep reviews and Scite for citation context.
FAQ#
Q: How does the consensus meter work?
It aggregates each retrieved study’s stance into a strength rating, not a simple majority vote.
Q: Can I trust the answer?
Cross-check the linked studies; the meter summarises, it does not adjudicate.
Q: Does it cover all fields?
Strongest in biomedical, nutrition and social sciences; weaker in niche engineering.