Elicit
Overview
Elicit is an AI research assistant built specifically for academic literature. Unlike general-purpose chat AI, Elicit searches 138+ million academic papers and 545,000 clinical trials using semantic search, and helps with summarization, data extraction, and paper screening. Its target users are researchers, graduate students, clinicians, and evidence-based decision makers.
Key Features
1. Semantic Academic Search
Elicit surfaces relevant research across 138M+ papers based on meaning, not exact keyword matches — surfacing specialized work that general chatbots miss.
2. Systematic Review
Elicit automatically screens thousands of papers against custom inclusion/exclusion criteria and explains each decision. Output is PRISMA 2020-compliant and fully auditable; across 994 Cochrane reviews it recorded 95% search recall, 99% full-text screening accuracy, and 96% extraction accuracy.
3. Sentence-Level Citations
Every AI-generated claim links back to the exact sentence in the source paper, so researchers can verify grounding instantly — a strong defense against hallucination.
4. Data Extraction Tables
Elicit extracts interventions, outcomes, sample sizes, and statistical significance across multiple papers into a comparable table — invaluable for meta-analysis and literature reviews.
5. Notebooks
A single "living page" that unifies search, summaries, and systematic review. Combine papers across multiple queries and your own uploads, chat with just the papers you select, create as many summaries as you need, and narrow down with filters and extraction columns straight into a systematic literature search.
6. Elicit Research Agent
Launched August 4, 2026 as an environment built for high-stakes decisions. It synthesizes evidence across scientific literature, clinical trials, patents, regulatory databases, and other sources, producing fully cited reports, analyses, tables, and visualizations — aimed at work like competitive landscape analysis that paper search alone cannot finish.
7. API and MCP Server
The Search API (138M papers, 545K clinical trials), Reports API (cited research answers), and Systematic Review API (end-to-end reviews) are all callable programmatically. The same capabilities are exposed through an MCP server, so Elicit runs directly inside Claude, ChatGPT, Gemini, and other MCP-compatible clients.
Pricing
| Plan | Monthly | Key Perks |
|---|---|---|
| Basic | $0 | Unlimited search, 2 automated reports/mo, 2 extraction columns |
| Plus | $12 ($120/yr) | 4 automated reports per month |
| Pro | $49 ($499/yr) | 12 reports or systematic reviews per month, unlimited high-accuracy columns |
| Team | $79/seat ($780/yr, 2-seat minimum) | 20 reports, collaboration, sharing |
| Enterprise | Contact sales | Org-level features |
Typical Use Cases
- Thesis and dissertation literature reviews
- Systematic reviews for medical and clinical research
- Policy and evidence-based decision support
- Patent and technical landscape scouting
- Running evidence gathering from inside Claude or ChatGPT over MCP
Strengths
- Purpose-built for 138M+ academic papers and clinical trials
- Sentence-level citations mitigate hallucination risk
- Dramatically speeds up systematic reviews
- API and MCP let you embed it in your own workflows
Weaknesses
- Weaker for non-academic content (general news, etc.)
- Monthly report limits can feel tight on heavy projects
- Team plan is expensive for small labs
- Non-English literature coverage lags English
Official
- Website: https://elicit.com