DOSSIER · Research assistant · FLAGSHIP · VERIFIED 13 Oct 2026
Consensus58FlagshipCited academic Q&A and claim-level literature scanningVerified 2026-10-13
Dossier · Research assistant
Consensus
Consensus flagship-ready dossier: Cited academic Q&A and claim-level literature scanning.
01
What is Consensus?
Decision
Cited academic Q&A and claim-level literature scanning
Choose it when
Fast yes/no or claim-shaped research questions that need peer-reviewed citations
Students who need a first-pass literature answer with links back to papers
Skip it when
You need a full systematic-review workflow with reproducible search strings, screening audit trails, and PRISMA documentation
You plan to cite summaries without opening the underlying papers
Cost
Free $0 · Pro $20/mo or $144/yr ($12/mo annual) · Deep $65/mo or $540/yr ($45/mo annual) · Teams/Enterprise custom · verified students and faculty: 40% off Pro for 12 months · clinicians: 25% off.
Privacy
Consensus privacy policy is dated August 18, 2026.
Watch for AI summaries can over-compress uncertainty
Consensus has a free plan with 3 Deep searches and 10 Pro message credits a month. Pro costs $20 a month or $144 a year. Verified students and faculty get 40% off Pro for 12 months. Pricing checked 13 October 2026.
Role: Cited academic Q&A and claim-level literature scanningCategory: Research assistantSynthesized
Consensus at a glance, with the date each field was checked.
Field
Value
Editorial fit
58 out of 100
Dossier status
Flagship
Set
Research assistant
Role in a workflow
Cited academic Q&A and claim-level literature scanning
Pricing
Free Free · Pro (monthly) $20/mo · Pro (annual) $144/yr · Deep (monthly) $65/mo · Deep (annual) $540/yr · Teams Custom · Enterprise Custom
Training on your data
Consensus says it does not use your personal or non-personal information to train large language models, or send user data to third parties for that purpose. You do not need to opt out of that stated use. Its privacy policy covers the website and mobile apps without different training rules for free, paid, team or enterprise accounts; it does not set out a separate API training rule. This is a promise about LLM training, not a promise that your searches are never stored or processed. Source: Consensus, Privacy policy (linked in Sources)
Last verified
2026-10-13
Editor's note
Consensus is useful when the user has a question rather than keywords. VerdictPal should praise fast, cited orientation while making the user verify the papers before turning an answer box into a claim.
Public facts · vendor & repo
List prices and pay-as-you-go entry points we can cite without running our own bench. Each tile links to a source when possible.
The public positioning for this product — the loop we score against on VerdictPal.
01
Start with the wedge
Fast yes/no or claim-shaped research questions that need peer-reviewed citations
02
Run the representative task
Ask 10 yes/no or claim-shaped literature questions in Consensus, Elicit, Google Scholar, and manual library search.
03
Check the failure modes
AI summaries can over-compress uncertainty
04
Compare before recommending
Compare against Elicit, Scite, Semantic Scholar before shipping advice.
02
Who it fits, where it fails
Best for
Fast yes/no or claim-shaped research questions that need peer-reviewed citations
Students who need a first-pass literature answer with links back to papers
Comparing paper-level evidence before deciding whether to run a deeper review
Users who want Research Agent style multi-step search planning and academic filters
Avoid if
You need a full systematic-review workflow with reproducible search strings, screening audit trails, and PRISMA documentation
You plan to cite summaries without opening the underlying papers
Your topic depends heavily on books, policy documents, grey literature, or sources outside its scholarly corpus
You need a completely transparent ranking/corpus methodology for final evidence grading
Strengths
AI academic search
cited answers
Research Agent
multi-step search planning
Pro Analysis
Ask Paper
advanced filters
paper lists
institution links
Zotero integration
incognito mode
history controls
Weaknesses
AI summaries can over-compress uncertainty
Retrieval/ranking can make a field look settled
Academic-paper corpus can miss books, policy, and niche sources
Agent workflows can still miss important search terms
UX may encourage citing answer text instead of papers
How it could improve
Summaries that flatten disagreement should preserve the debate, present multiple positions and mark where the literature is contested instead of smoothing it into one answer.
Sparse coverage in niche fields is a corpus problem, not an algorithm one. Consensus should surface a coverage-confidence indicator so the reader knows when results are thin rather than pretending otherwise.
Metadata quality could be improved with a verification layer that cross-checks imported records against authoritative databases before they enter the library.
Privacy posture scores 40/100, the weakest dimension on this dossier. Addressing this would meaningfully raise the editorial fit.
Search modes · 4 lenses
One input box, many retrieval postures. Filter by tier.
01Free
Default
Default path for Consensus — verify limits on the live product.
02Pro
Deep work
Deep work path for Consensus — verify limits on the live product.
03Pro
Focused task
Focused task path for Consensus — verify limits on the live product.
04Free
Export & share
Export & share path for Consensus — verify limits on the live product.
Competitive lens
Where Consensus wins for cited research — and where a rival still belongs in the stack.
Consensus is stronger when fast yes/no or claim-shaped research questions that need peer-reviewed citations; Elicit may still win for narrower fit, procurement, or specialist depth.
Consensus
Fast yes/no or claim-shaped research questions that need peer-reviewed citations
Students who need a first-pass literature answer with links back to papers
Elicit
You need a full systematic-review workflow with reproducible search strings, screening audit trails, and PRISMA documentation
You plan to cite summaries without opening the underlying papers
03
Scores and evidence
Metric lab · 16 dimensions
Click a tile for the editorial note. Color follows score: coral, yellow, mint.
Shape
Avg 69 · 40–98
58Editorial fit
Weighted roll-up across 13 dimensions for Research assistant; pending desk verification if rescored from public facts.
By group
Fit78
Editorial fit58
Wedge task fit82
Feature depth94
Cost67
Free-tier utility71
Cost-to-value70
Opportunity cost59
Trust70
Source grounding58
Privacy posture40
Failure transparency98
Evidence strength55
Transparency98
Workflow64
Integration reach60
Setup friction58
Reliability70
Competitive position75
Data portability59
Benchmark ledger
Public rows are vendor or third-party claims we logged with a date. Desk rows are reserved for VerdictPal self-run results.
Measure
Result
Source
Ask 10 yes/no or claim-shaped literature questions in Consensus, Elicit, Google Scholar, and manual library searchsource relevance, answer caution, paper verification effort, missing seminal studies, uncertainty preservation.
Planned
VerdictPal benchmark plan
Use Research Agent for 5 complex multi-step research questionssearch planning quality, filter use, citation support, missed branches, reproducibility.
Planned
VerdictPal benchmark plan
Test incognito mode, Zotero integration, institution links, history clearing, and account deletiondiscoverability, actual workflow value, privacy friction, export/lock-in.
Planned
VerdictPal benchmark plan
Editorial evidence · 0 entries
No editorial benchmark yet. The tool stays at status Flagship until evidence lands.
This tool is in the VerdictPal Citation Fidelity protocol — frozen question bank, dual-reviewer grading, results still pending. Citation Fidelity v0.1 →
Consensus remains the claim-shaped academic Q&A tool. It is strongest when the user has a research question and needs cited orientation quickly. This pass adds current corpus/user-scale claims, Research Agent, account/privacy controls, institutional full-text linking, and a sharper “answer box is not a literature review” warning.
Converted imported research into metrics, panes, scenarios, benchmark rows, comparison notes, and pricing deck.
04
What it costs, what it keeps
Pricing deck · checked 2026-10-13
Desk pick
Pro (monthly)
$20
month
Imported from the current pricing summary.
Verify live regional checkout before publishing procurement advice.
Deep (monthly)
$65
month
Imported from the current pricing summary.
Verify live regional checkout before publishing procurement advice.
RAW · checked 2026-10-13: Free $0 · Pro $20/mo or $144/yr ($12/mo annual) · Deep $65/mo or $540/yr ($45/mo annual) · Teams/Enterprise custom · verified students and faculty: 40% off Pro for 12 months · clinicians: 25% off.
Privacy deep-dive · checked 2026-09-14
Consensus privacy policy is dated August 18, 2026. The policy says the service encrypts data in transit and at rest, does not sell personal information, and does not use user data to train LLMs including third-party models. Queries and uploads are processed by model providers (OpenAI, Anthropic, Google, Baseten) as subprocessors without account identifiers. Account settings include thread-history clearing and account deletion.
Training on your dataConsensus says it does not use your personal or non-personal information to train large language models, or send user data to third parties for that purpose. You do not need to opt out of that stated use. Its privacy policy covers the website and mobile apps without different training rules for free, paid, team or enterprise accounts; it does not set out a separate API training rule. This is a promise about LLM training, not a promise that your searches are never stored or processed. Source: Consensus, Privacy policy (linked in Sources)
EU data residencyUnknown
SOC 2 attestationUnknown
Local-first by defaultUnknown
Under the hood
Vendors named on the product about page — useful for procurement and privacy reviews.
Editorial lenses only. Subscription and API pricing live in their own sections above.
Lens 01
Positioning
Consensus is framed here as Cited academic Q&A and claim-level literature scanning. Consensus remains the claim-shaped academic Q&A tool. It is strongest when the user has a research question and needs cited orientation quickly. This pass adds current corpus/user-scale claims, Research Agent, account/privacy controls, institutional full-text linking, and a sharper “answer box is not a literature review” warning.
Checked
2026-06-07
Evidence
synthesized
Fast yes/no or claim-shaped research questions that need peer-reviewed citations
Students who need a first-pass literature answer with links back to papers
Comparing paper-level evidence before deciding whether to run a deeper review
Users who want Research Agent style multi-step search planning and academic filters
Promoted to flagship depth for the student research shelf: the dossier already carries the full template; pricing checked 2026-09-14. Benchmark status stays pending.
Live-verified Consensus pricing and privacy (2026-09-14). Individual: Free $0; Pro $20/mo or $144/yr; Deep $65/mo or $540/yr. Discount wording corrected 13 Oct 2026: verified students and faculty get 40% off Pro for 12 months; clinicians get 25% off. Privacy policy dated August 18, 2026: encrypt in transit/at rest; no sale of personal information; no LLM training including third-party models.
Consensus remains the claim-shaped academic Q&A tool. It is strongest when the user has a research question and needs cited orientation quickly. This pass adds current corpus/user-scale claims, Research Agent, account/privacy controls, institutional full-text linking, and a sharper “answer box is not a literature review” warning.
Converted imported Notion research into a full flagship-ready dossier template with metrics, panes, pricing deck, scenarios, benchmark rows, and comparison slices.
Every answer below is assembled from the dated fields on this page. Nothing is written separately for search.
Is Consensus AI free?
Consensus has a free plan with 3 Deep searches and 10 Pro message credits a month. Pro costs $20 a month or $144 a year. Verified students and faculty get 40% off Pro for 12 months. Pricing checked 13 October 2026.
What does the free Consensus plan include?
The free plan includes 3 Deep searches and 10 Pro message credits a month. A Deep search also uses one Pro message credit, according to Consensus's documentation.
How much is Consensus Pro?
Consensus Pro costs $20 per month or $144 per year and includes unlimited Pro messages and 15 Deep searches a month. Deep costs $65 per month or $540 per year and includes 200 Deep searches a month.
Is there a Consensus student discount?
Students and faculty get 40% off after verifying an academic email, with manual review available if needed. Consensus's pricing file lists the Pro student discount for 12 months. Clinicians get 25% off after NPI verification.
Does Consensus train AI on my data?
Consensus's privacy policy says queries, content and personal information are never used to train its own or third-party AI models. Some processing uses OpenAI, Anthropic, Google and Baseten as service providers; processing is not the same as training.
Is Consensus worth using?
Consensus scores 58 out of 100 on editorial fit and carries Flagship dossier status. Its job in a research workflow is: Cited academic Q&A and claim-level literature scanning.
Who is Consensus best for?
Consensus earns its place when you need:
Fast yes/no or claim-shaped research questions that need peer-reviewed citations
Students who need a first-pass literature answer with links back to papers
Comparing paper-level evidence before deciding whether to run a deeper review
Users who want Research Agent style multi-step search planning and academic filters
When should you not use Consensus?
Skip Consensus in these cases:
You need a full systematic-review workflow with reproducible search strings, screening audit trails, and PRISMA documentation
You plan to cite summaries without opening the underlying papers
Your topic depends heavily on books, policy documents, grey literature, or sources outside its scholarly corpus
You need a completely transparent ranking/corpus methodology for final evidence grading
What goes wrong with Consensus?
The dossier publishes 5 failure modes for Consensus, and they stay published whether or not the vendor likes them:
AI summaries can over-compress uncertainty
Retrieval/ranking can make a field look settled
Academic-paper corpus can miss books, policy, and niche sources
Agent workflows can still miss important search terms
UX may encourage citing answer text instead of papers
What are the alternatives to Consensus?
The closest options to Consensus are Elicit, Scite, Semantic Scholar, Perplexity, Connected Papers. Each one that has a tool in the atlas is linked from this dossier, with a head-to-head comparison.
The Pack · Editorial newsletter
Know when the rules move.
One short email when a university changes its AI rule for theses, or a research tool changes its pricing or privacy terms. No tracking, no third-party analytics. Unsubscribe in one click.