# Elicit — An AI research agent built for scientific literature, search, screen and extract cited data from 138M+ papers.

> Source: The Agents Index — https://theagentsindex.com/elicit (structured, researched, re-verified)
> Facts last verified: 2026-07-23

Elicit is an AI research agent purpose-built for scientific and academic literature, not the open web. It began in 2021 inside Ought, a nonprofit co-founded by CEO Andreas Stuhlmüller to research AI reasoning, and spun out as an independent public benefit corporation, raising a $9M seed round and later a $22M Series A (Spark Capital, Footwork) at a $100M valuation. Given a research question, its Research Agent autonomously plans sub-questions, searches over 138 million papers, and extracts sourced findings into structured tables — used by 5M+ researchers across academia, pharma, policy and industry.

| Fact | Value |
| --- | --- |
| Website | https://elicit.com |
| API | Yes |
| Best for | Academic researchers, R&D and policy teams, and students running literature reviews or formal systematic reviews who need cited, structured evidence pulled from published research. |
| Not for | Buyers wanting a general web-search answer engine (see Perplexity) or a broad autonomous agent that automates open-web tasks and deliverables beyond published literature (see Manus), Elicit is scoped to the academic/scientific literature, not the open web. |

## Verdict

Elicit is the research agent in this category actually built for scientific literature rather than the open web. Its Research Agent autonomously plans and runs a multi-step literature investigation, its Systematic Review workflow automates the screening step of a formal review at real scale (thousands of papers, PRISMA-grade per its own claim at the top tier), and every generated claim links back to a sentence-level citation in the source paper — a genuinely different job than Perplexity's live-web answers or Manus's general task automation. The honest caveats are self-disclosed: Elicit's own docs admit hallucination risk and tell users to verify important findings against the original paper, independent reviewers find it can miss papers a manual search would catch, and it performs best on empirical, formulaic research rather than open-ended theoretical work. Pick Elicit if you need cited, structured evidence from published literature at academic rigor; skip it if you want general web research or broader task automation.

## Strengths and weaknesses

- ✓ A genuine Systematic Review workflow that automates screening at scale (5,000-40,000 papers depending on tier) — a real PRISMA-adjacent process, not a chat wrapper over search.
- ✓ Every generated claim carries a sentence-level citation back to the source paper, and the Research Agent extracts structured data rather than just summarising.
- ✓ Backed by real scale and continuity: 5M+ researchers, $31M total raised ($9M seed + a $22M Series A led by Spark Capital), and a public-benefit-corporation structure aligned with its research mission.
- ✗ Independent reviewers note it can miss papers a traditional database search would catch, especially very recent ones — for a formal systematic review it should supplement, not replace, comprehensive manual retrieval.
- ✗ Elicit's own documentation acknowledges hallucination is a real risk despite its safeguards, and directs users to verify important findings against the original paper rather than trust the summary alone.
- ✗ It is strongest on empirical, formulaic research (clinical trials, ML benchmarks); less formulaic or open-ended theoretical work is more likely to lose nuance.
- ✗ Pro pricing ($49/mo) is steep next to narrower single-purpose competitors like Consensus (roughly $9-10/mo) if all you need is quick paper lookups rather than full systematic-review tooling.
- ⚠ Best treated as a supplement to, not a replacement for, exhaustive manual search in a formal systematic review.
- ⚠ Hallucination risk is real and self-disclosed by Elicit, verify load-bearing findings against the source paper.

## Key features

- **Research Agent** — Autonomously plans sub-questions, searches the literature and assembles findings into an interactive, cited report or table.
- **Systematic Review workflow** — Automates screening thousands of papers against inclusion criteria for a formal literature review, up to 40,000 papers on Enterprise.
- **Structured data extraction** — Pulls specific data points (sample size, method, outcome) out of papers into a table with up to 40 columns, not just a summary.
- **138M+-paper search with citations** — Every generated claim links back to a sentence-level citation in the source paper so you can check it yourself.
- **Zotero import + API** — Bring an existing reference library in, and (Pro tier and above) call Elicit's search and extraction layer from your own code.

## Use cases

- **Literature review for a new project** — A researcher starting a new study asks Elicit to map what has already been published on a question before designing their own work.
- **Formal systematic review** — A team runs the Systematic Review workflow to screen thousands of candidate papers against inclusion criteria for a PRISMA-style review.
- **Cross-paper data extraction** — A policy or pharma analyst extracts the same handful of data points (sample size, effect size, population) across dozens of papers into one comparable table.

## Integrations

Zotero (import an existing library) · API access (Pro tier and above) · Extraction from 135-200 external data sources depending on tier · Model Context Protocol (MCP) — Elicit ships an official MCP server so Claude Desktop, Claude Code and other MCP clients can call its search/extraction API

## Sources

- https://elicit.com/pricing
- https://elicit.com
- https://elicit.com/team
- https://elicit.com/blog/series-a/
- https://support.elicit.com/en/articles/552897
- https://docs.elicit.com/