Written and maintained by CASRAI Editorial Board
Published · Last updated
On this page
Consensus (consensus.app) is a named commercial AI-powered search engine for peer-reviewed research papers. Given a natural-language question, it retrieves papers from an indexed academic corpus and generates a synthesis of what those papers report, and for questions phrased as a testable claim it displays a Consensus Meter — a visual breakdown of how the retrieved papers agree, disagree, or split on that claim. It is one entry in CASRAI’s AI-powered research assistant tools guide, alongside comparable tools such as Elicit, SciSpace, AnswerThis, and Undermind. This guide covers what Consensus actually does, how the Consensus Meter works and where it can mislead, how to verify its output, what it is and is not suited for, published pricing, and how it compares to the alternatives researchers most often weigh it against.
What Consensus is, and what it isn’t
Consensus is not a generic descriptor for “AI that searches papers” — it is a specific product built by a company also called Consensus. It sits between three different tools researchers already know, and understanding those boundaries is most of what you need to use it well:
- Versus keyword search (a library database, PubMed, Google Scholar): keyword search returns a ranked list of documents matching your terms; you still have to read each one and synthesize the answer yourself. Consensus adds a synthesis layer on top of retrieval — it reads the retrieved papers and generates a written summary of what they say, with each point in that summary linked back to a specific source paper.
- Versus a general-purpose AI chatbot: a chatbot with no dedicated retrieval step against an indexed academic corpus can be asked to “summarize the research,” but its answer is not reliably grounded in retrievable, checkable sources, and carries a materially higher risk of fabricated or unverifiable citations. Consensus’s core function is running the retrieval step first, against a defined corpus of papers, and then synthesizing only across what it actually retrieved.
- Versus a systematic review: this is the comparison that matters most and the one vendor marketing is least likely to volunteer — see the Consensus Meter section below.
How does Consensus AI work?
Consensus AI works in three steps: it searches an academic index for papers matching your question, ranks them, and then summarizes the top results with each point linked to a source paper. For yes/no questions it also shows the Consensus Meter, a tally of how many retrieved papers say yes, no, mixed or possibly. It is built for exploratory, question-led research, and the company itself says it is not designed to produce evidence syntheses or systematic reviews (Deakin University Library evaluation).
- Retrieval. A hybrid of keyword and embedding-based semantic search. In one independent walkthrough the first pass pulls roughly 1,500 candidate papers (Aaron Tay’s 2025 deep dive).
- Re-ranking. Candidates are re-ranked by how well they match the query and by “research strength” signals such as citation counts, recency and journal reputation, according to the same review.
- Summarization depth. The review reports that Quick mode summarizes up to 10 papers, Pro up to 20 and Deep up to 50.
- Mostly abstracts. For most paywalled papers Consensus is working from titles and abstracts only; full text is mainly available for open-access papers, so details that appear only in methods or results sections can be missed.
- AI-extracted labels. Study-design tags and similar details are generated by AI and can be wrong, and LLM-based pipelines may return slightly different rankings on repeat runs.
What corpus it actually searches
Consensus says its index covers more than 400 million scholarly sources, drawn from Semantic Scholar, OpenAlex, its own crawl, publisher partnerships (it states that six of the twelve largest publishers participate) and all of PubMed (Consensus, how Consensus works). Vendor figures are not consistent over time: older Consensus pages cite 150 million and 200 million sources, and an independent 2025 review put the corpus at roughly 200 million records, mainly from OpenAlex and Semantic Scholar (Aaron Tay). Treat any count as a dated vendor claim rather than an audited number.
The practical point is different from the count: Consensus is not searching a single curated collection the way a systematic-review protocol specifies a fixed set of databases (for example MEDLINE, Embase, CENTRAL) with a documented search date. It queries a broad, general aggregator, which is a different and less reproducible evidentiary basis. For a research database with its own AI features, see how Dimensions AI works.
The search pipeline step by step
Consensus describes a pipeline in which search comes before any generative AI (how Consensus works):
- Hybrid retrieval. Embedding-based semantic search is combined with BM25 keyword matching over titles and abstracts.
- First re-ranking. The top 1,500 candidates are re-ranked using relevance, recency, citation counts and journal reputation.
- Second re-scoring. The top 20 are re-scored with a larger model.
- Generation. Only then do language models summarize. Consensus states that it uses commercial models (OpenAI is named) for summaries and fine-tuned open-source models for the Consensus Meter, and that “checker models” verify relevance.
Consensus also states that user data is not used for model training. Because relevance is judged mainly from titles and abstracts, a paper whose key result appears only in the full text can be ranked or summarized on incomplete information.
Search modes and Deep Search
An independent 2025 review describes three depths of summarization (Aaron Tay):
| Mode | Papers summarized |
|---|---|
| Quick | Up to 10 |
| Pro | Up to 20 |
| Deep | Up to 50 |
In the reviewer’s worked Deep Search example, the tool ran 20 queries returning 1,004 hits plus 43 from citation searching, filtered these to 573, and summarized the top 50. Most papers were read from the abstract; full text was used for open-access papers and some publisher-partnership content. Rankings varied slightly between runs, and the tool passed a “negative rejection” test, meaning it declined to answer from unrelated papers. The same review notes that another tool, Undermind, evaluates about 150 results on average, a useful contrast when you need breadth. The data table below the guide lists how many Deep reviews each paid plan allows.
The Consensus Meter — what it aggregates, and its real epistemic limits
This is the part of the product most worth being plain-spoken about, because Consensus’s own marketing has no incentive to be. For a question phrased as a testable yes/no/maybe claim (for example, whether a specific intervention affects a specific outcome), Consensus can display a Consensus Meter: a bar showing how many of the papers it retrieved and summarized for that specific query support the claim, oppose it, or report mixed results.
What the meter is actually counting is votes across a heterogeneous, retrieval-dependent sample of studies — not a weighted synthesis of evidence quality. That distinction matters more than it sounds:
- It does not weight by study design. A well-powered randomized controlled trial and a small observational study or an animal model each count as one paper toward the bar, with no explicit weighting for where each sits in a standard evidence hierarchy.
- It does not weight by sample size or statistical power. A study with an n of 20 and a study with an n of 20,000 can each contribute one “supports” or “opposes” vote.
- It does not apply a risk-of-bias assessment. Tools like the Cochrane risk-of-bias tool exist specifically because study quality varies enormously even among published, peer-reviewed papers on the same question; the meter has no equivalent appraisal step.
- It is sensitive to how the query is retrieved and phrased. The meter reflects agreement only within the specific set of papers Consensus’s retrieval step happened to surface for that exact query on that day — a differently worded question, or a query run a month later against an updated index, can surface a different paper set and a different bar.
- It does not report effect sizes. A bar full of “supports” tells you direction of findings, not magnitude, clinical/practical significance, or confidence intervals — a claim can be technically “supported” by five studies each reporting a trivially small effect.
Consensus describes the mechanics in its own words (Consensus Meter announcement, January 2023): the meter launched in beta on 1 February 2023 and works only on yes/no questions. A language model reads the first 20 search results and classifies each as yes, no or possibly, dropping results that are not relevant or have low classification confidence. The supporting text shown for each paper is extracted word for word. The team estimates roughly 10% misclassification, worse with double negatives, and states that the meter does not weight study quality, so a meta-analysis counts the same as a case report. Misclassifications can be reported to the address given on that page.
In short: vote-counting across heterogeneous studies is not evidence synthesis. A systematic review under a framework like PRISMA 2020 exists precisely to control for these problems — a documented, reproducible search strategy against fixed databases, explicit inclusion/exclusion criteria, formal risk-of-bias appraisal of every included study, and (in a meta-analysis) a statistically weighted pooled effect estimate, not a raw count of papers on each side. A Consensus Meter bar that looks decisively one-sided can be resting on a handful of small, low-quality, or methodologically weak studies, and there is no way to tell that from the bar itself — you have to open the underlying papers and read them. Use the meter as a fast orientation signal for scoping a question, never as a stand-in for critical appraisal or a substitute for the judgment a systematic review is built to provide.
How to verify what Consensus tells you
Every claim in a Consensus synthesis and every paper counted in a Consensus Meter is meant to be traceable to a specific source. Before citing or acting on anything Consensus produces:
- Open the underlying paper, not just the AI-generated snippet. Confirm the paper actually says what the synthesis claims it says — AI summarization of scientific findings can flatten nuance, drop qualifiers (“in this specific population,” “under these conditions”), or misstate direction of effect.
- Check the study design and sample size yourself. Since the meter doesn’t surface this, you have to look: is this a randomized trial, an observational study, a case report, an animal or in-vitro study, a preprint that hasn’t been peer reviewed?
- Check publication status and venue. A preprint and a paper published in a high-quality peer-reviewed journal after review should not carry equal weight in your own judgment, even though they may both show up as a single “supports” vote.
- Re-run the query with different phrasing. If the meter shifts meaningfully with how the question is worded, that instability is itself informative — it tells you the underlying evidence base is thin or the claim is not cleanly testable with a single retrieval query.
- Cite the original paper, not the Consensus synthesis. In a manuscript, grant proposal, or evidence brief, the citation belongs to the peer-reviewed source you verified, not to Consensus’s summary of it.
What Consensus is genuinely good for
- Fast orientation on an unfamiliar question. Getting a first read on whether a claim has any published support at all, and a starting set of papers to read, before committing time to a deeper search.
- Sanity-checking a specific claim. Quickly checking whether a statement in a draft, a proposal background section, or someone else’s argument has any peer-reviewed backing, as a first pass before deeper verification.
- Scoping a topic before designing a formal search. Seeing roughly how much literature exists and which terms and subtopics come up, to inform a more rigorous search strategy later.
- Non-systematic narrative writing. Background sections, blog-style science communication, or grant narrative where a fast, source-linked orientation is useful and the writing does not claim to be an exhaustive or reproducible synthesis.
What it is genuinely bad for
- Anything protocol-driven. A registered systematic review, a Cochrane review, a health-technology assessment, or any output that requires a documented, reproducible, pre-specified search strategy against named databases. Consensus’s retrieval is not reproducible in that sense and does not log a fixed search date and query against a defined database set the way review protocols require.
- Clinical, regulatory, or high-stakes decisions. The Consensus Meter’s vote-count is not risk-of-bias-adjusted evidence and should never substitute for guideline-level or clinician-appraised evidence.
- Meta-analysis or effect-size questions. “Does X work” and “how large is X’s effect, and how confident are we” are different questions; Consensus answers something closer to the first, imperfectly, and does not answer the second at all.
- Exhaustive or auditable literature coverage. Because the retrieved paper set is not fixed or fully disclosed, two people running the same query cannot be guaranteed to see the same result set indefinitely, which is disqualifying for any use case that requires an auditable search trail.
Why it does not replace a PRISMA-registered search
A systematic review conducted under PRISMA 2020 reporting standards documents an explicit, reproducible search strategy across named databases, a defined search date, explicit inclusion/exclusion criteria applied by (typically) two independent reviewers, formal risk-of-bias assessment of every included study, and — where a meta-analysis is warranted — a statistically pooled, weighted effect estimate with a stated confidence interval. Consensus’s retrieval-and-synthesis process is not designed to produce any of that: the corpus queried is a broad aggregator rather than a fixed set of named databases, the search is not logged as a reproducible protocol, inclusion is not governed by explicit criteria applied consistently by independent reviewers, and there is no risk-of-bias appraisal step. None of that makes Consensus a bad tool — it makes it a different one, built for fast orientation rather than for the auditability a formal evidence synthesis requires. Treat a Consensus Meter result as a hypothesis to check, never as a completed synthesis.
How to use Consensus AI for research
A workable routine treats Consensus as the first step of a search, not the last. It pairs naturally with a well-formed research question and a method-aware AI-assisted literature review workflow.
- Phrase a specific question. A question with a clear population, intervention and outcome retrieves tighter results than a topic label. Yes/no framing is what triggers the Consensus Meter.
- Read the top papers, not only the summary. Open each linked paper and check design, sample and venue as described above.
- Reword and re-run. If the meter or the paper set changes a lot, the evidence is thin or the question is too loose.
- Move to a reproducible search for anything that will be published as a review. See AI tools for systematic literature reviews for what AI can and cannot do at the screening and extraction stages.
- Cite the original papers you verified, never the Consensus summary.
If you are choosing between Consensus and a general answer engine, the comparison in Perplexity vs. Consensus for research questions sets out the trade-offs.
Pricing and usage limits
Consensus lists five tiers in its subscription-plans help article: Free, Pro, Deep, Teams and Enterprise (Consensus help center). The article states these monthly allowances and prices; the data table below the guide compares them side by side.
- Free: $0, with 3 Deep reviews, 10 Pro messages, 30 API and MCP calls and 10 Study Snapshots per month.
- Pro: $20 per month, or $12 per month equivalent on annual billing ($144 per year); 15 Deep reviews, unlimited Pro messages and Study Snapshots, and 500 API and MCP calls per month.
- Deep: $65 per month, or $45 per month equivalent on annual billing ($540 per year); 200 Deep reviews, unlimited Pro messages and Study Snapshots, and 2,000 API and MCP calls per month.
- Teams: custom pricing; 50 Deep reviews per user per month, 500 API and MCP calls, and all Pro features.
- Enterprise: listed as a separate tier.
Plan structures and prices change, so confirm them on consensus.app before budgeting. Note that a plan’s API and MCP allowance matters if you intend to connect Consensus to other tools; the free allowance of 30 calls per month suits only testing.
Consensus app, website and common misspellings
“Consensus app” and “Consensus AI” refer to the same product, which runs in the browser at consensus.app. People often search for it as “concensus ai”, “consenus ai” or “consus ai”, and sometimes as “census ai research”. The product has no connection to the US Census Bureau or to census data: it searches published scientific literature. If a different tool appeared in your results, check that the address is consensus.app.
How Consensus compares to the alternatives
Researchers most often weigh Consensus against these tools. All are AI-assisted, but they solve different problems.
| Tool | Core approach | Where it’s strongest |
|---|---|---|
| Consensus | Retrieval + claim-level synthesis with the Consensus Meter agreement bar | Fast orientation on whether a specific claim has published support |
| Elicit | Structured evidence extraction into a comparable table (methods, outcomes, sample size) across a paper set, plus systematic-review-screening support | Building a structured, exportable evidence table; supporting (not replacing) the screening stage of a real review |
| SciSpace | Paper-level AI reading assistant — chat with a specific PDF, plus a literature-review/Discovery mode for broader search | Deep, close reading and question-answering against one paper or a small set you’ve already found |
| AnswerThis | Research-gap identification — surfaces what a body of literature has not yet answered | Early-stage proposal and thesis scoping, finding an unaddressed angle |
| Undermind | Iterative “deep search” agent that reasons across multiple search rounds before returning a curated set | Broad, exhaustive-feeling discovery for a complex, multi-faceted question |
| Semantic Scholar | Free scholarly search engine and citation graph, no AI-generated synthesis by default | Straightforward discovery, citation tracking, and as the underlying index several AI tools (including Consensus) build on |
| Scite | “Smart Citations” — classifies each citing paper as supporting, contrasting, or mentioning the cited claim, based on citation context rather than a synthesized meter | Checking how a specific published claim or paper has actually been treated by the papers that later cited it |
The distinction worth holding onto: Elicit, SciSpace and Undermind are built to help you assemble and read a paper set; Scite is built to show you how a specific claim has been cited afterward; Consensus is built to give you a fast, claim-level directional read across whatever it retrieves. None of the seven substitutes for a registered systematic review, and none of their vendors will tell you that as directly as this page does.
Frequently Asked Questions
Is Consensus AI accurate?
Consensus links every claim in its synthesis back to a specific retrieved paper, which makes it checkable — but “checkable” is not the same as “verified.” The accuracy of what it retrieves and how it characterizes each paper still needs to be confirmed against the source, and the Consensus Meter’s agreement bar should never be read as a validated accuracy signal about the underlying science; it only reflects the paper set retrieved for that specific query.
What is the Consensus Meter, exactly?
A visual bar, shown for questions phrased as a testable yes/no/maybe claim, indicating how many of the retrieved and summarized papers support, oppose, or report mixed findings on that claim. It is a vote count across a retrieval-dependent paper set, not a study-quality-weighted synthesis — it does not account for study design, sample size, risk of bias, or effect size.
Can I use Consensus for a systematic review or meta-analysis?
No. A registered systematic review requires a documented, reproducible search strategy across named databases, explicit inclusion/exclusion criteria, independent dual screening, formal risk-of-bias appraisal, and (for a meta-analysis) a statistically weighted pooled effect estimate. Consensus’s retrieval is not reproducible or auditable in that sense, and its meter performs none of the quality-weighting a review requires. Use it to scope a question before you design a review, not to conduct one.
Is Consensus free?
Yes, there is a free plan. According to the vendor’s subscription-plans help article it includes 3 Deep reviews, 10 Pro messages, 30 API and MCP calls and 10 Study Snapshots per month. Paid Pro and Deep plans raise those limits; see the pricing section above.
Does Consensus have an API?
The plan listing includes monthly API and MCP call allowances: 30 on Free, 500 on Pro and Teams, and 2,000 on Deep.
How many papers does Consensus search?
Consensus states that its index holds more than 400 million scholarly sources. Older vendor pages and an independent 2025 review give lower figures (150 to 200 million), so treat the number as a dated vendor claim.
Does Consensus read the full text of papers?
Mostly not. An independent review reports that most papers are processed from titles and abstracts, with full text for open-access papers and some publisher-partnership content.
What is Consensus Deep Search?
The deepest search mode, which summarizes up to 50 papers (Quick: 10, Pro: 20) after running multiple queries and filtering the candidate set.
What database does Consensus search?
Consensus says its search draws on Semantic Scholar, OpenAlex, its own crawl and publisher partnerships, and spans more than 400 million scholarly sources, aggregated across publishers, preprint servers and conference proceedings. It is a broad, continuously updated aggregator, not a fixed, named set of databases with a documented search date the way a systematic review protocol specifies.
How is Consensus different from ChatGPT or another general AI chatbot?
A general chatbot with no dedicated academic-retrieval step can be asked to summarize research, but its answer is not reliably grounded in an indexed corpus of actual papers and carries a materially higher risk of fabricated citations. Consensus runs a retrieval step against a defined academic corpus first, then synthesizes only across what it actually retrieved, with each point traceable to a source paper.
Sources
Sources:
Data: Consensus AI: usage limits and prices by plan, 2026
Compares Consensus plans by monthly price and monthly allowances for Deep reviews, Pro messages, API and MCP calls and Study Snapshots, compiled from the vendor's subscription-plans help article.
Period covered: 2026.
| Plan | Price, monthly billing (USD per month) | Price, annual billing (monthly equivalent) (USD per month) | Deep reviews allowed (per month) | Pro messages (per month) | API and MCP calls (per month) | Study Snapshots (per month) |
|---|---|---|---|---|---|---|
| Free | 0 | 0 | 3 | 10 | 30 | 10 |
| Pro | 20 | 12 | 15 | Unlimited | 500 | Unlimited |
| Deep | 65 | 45 | 200 | Unlimited | 2000 | Unlimited |
| Teams (per user) | Custom | Custom | 50 | All Pro features | 500 | All Pro features |
Sources
- Consensus subscription plans (help center) (accessed 2026-11-16)
Compiled by CASRAI from the sources above. Free to reuse under CC BY 4.0 with attribution.







