Ask a question. Get the scientific consensus. Instantly.
Consensus searches over 200 million academic papers to answer scientific questions with evidence-based responses. Each answer includes relevant papers, a consensus meter showing the strength of scientific agreement, and AI-synthesized summaries — making peer-reviewed research accessible without academic training.
Consensus is an AI search engine for scientific knowledge — designed to answer questions the way a scientist would, by consulting the research literature. Ask any empirical question and Consensus searches 200M+ peer-reviewed papers to find the most relevant evidence, generates an AI-synthesized answer, and provides a visual consensus meter showing how strongly the evidence supports the claim. Unlike general AI assistants that may confidently state uncited claims, Consensus grounds every answer in the research literature with paper citations. It's used by clinicians, policy analysts, journalists, and anyone who wants evidence-based answers to scientific questions without needing to be a researcher. The Copilot feature enables deeper synthesis and analysis of the evidence.
Before accepting or publishing a scientific claim ('Does X cause Y?', 'Is intervention Z effective?'), run it through Consensus. The consensus meter instantly shows how strongly the evidence supports or contradicts the claim, with the supporting papers cited.
Quickly check the current state of evidence on a treatment, diagnostic approach, or clinical question. Consensus synthesizes the relevant trials and studies, shows the strength of the evidence base, and surfaces the key papers without requiring a full PubMed search.
For any proposed intervention or policy question ('Does minimum wage increase unemployment?', 'Do school uniforms improve behavior?'), Consensus provides the current state of research evidence with the consensus meter, enabling evidence-informed policy analysis.
The Consensus Meter is a visual indicator that shows the overall strength of scientific agreement on the question you asked — ranging from 'Highly Supported' to 'Contradicted' to 'Insufficient Evidence.' It's generated by analyzing the direction and strength of findings across the papers Consensus retrieves. It provides instant context on how settled the science is before you read individual papers.
Google Scholar is a keyword search engine for papers — you get a list of results to click through manually. Consensus is an AI that answers your question using the papers — it reads and synthesizes across results to give you a direct answer with a consensus meter, instead of making you do the synthesis yourself. Consensus is better for questions; Google Scholar is better for finding a specific paper you already know exists.
Yes — Consensus is widely used by clinicians for rapid evidence checking. However, it should supplement, not replace, clinical judgment and thorough literature review for clinical decisions. The consensus meter and synthesized summaries are useful for quick evidence checks; for comprehensive clinical systematic reviews, Elicit or a formal database like Cochrane is more appropriate.
The gold standard for AI research — 50-100 sources, one cited report.
View Review & Details →Google's deep research — free tier available, powered by Gemini 2.5 Pro.
View Review & Details →Fast, cited deep research — free tier available with 5 queries per day.
View Review & Details →