Back
Elicit Solutions - Systematic Review
webCredibility Rating
3/5
Good(3)Good quality. Reputable source with community review or editorial standards, but less rigorous than peer-reviewed venues.
Rating inherited from publication venue: Elicit
Elicit is a practical research tool useful for AI safety researchers conducting literature reviews or evidence synthesis, but is not directly focused on AI safety topics itself; its relevance is methodological rather than substantive.
Metadata
Importance: 25/100tool pagetool
Summary
Elicit is an AI research tool designed to accelerate systematic literature reviews by automating key steps including title/abstract screening, data extraction, and evidence synthesis. It claims 80% time savings on screening, 96.4% screening recall, and 94% data extraction accuracy across a corpus of 138 million papers.
Key Points
- •Automates title/abstract screening using AI, with per-criterion scoring and rationale with supporting quotes from papers
- •Claims highest data extraction accuracy among AI products, supporting both quantitative and qualitative data including tables
- •Covers 138 million papers with semantic search across multiple databases to reduce search overlap
- •Generates structured reports with PRISMA diagrams, paper summaries, and sentence-level citations, though writing is not fully automated
- •Allows iterative refinement of screening criteria on pilot sets before full deployment, improving protocol design
Cited by 1 page
| Page | Type | Quality |
|---|---|---|
| Elicit (AI Research Tool) | Organization | 63.0 |
Cached Content Preview
HTTP 200Fetched Mar 20, 20269 KB
[Skip to main content](https://elicit.com/solutions/systematic-review#)
**AI for evidence synthesis**
# Systematic Literature Reviews
## Use AI to synthesize more evidence faster
[Try now](https://elicit.com/users/auth?show=signup)
See those asterisks? They’re magical quotation tools. [Try them for yourself →](https://elicit.com/users/auth?show=signup)
## Elicit's systematic reviews at a glance
80%
Time savings, especially for screening & data extraction
96.4%
Up to 96.4% screening recall
94%
Data extraction accuracy; up to 99.4% in some cases
138m
Papers to include in your review
[Latest update\\
\\
How we evaluated Elicit Reports\\
\\
Read more](https://elicit.com/blog/elicit-reports-eval)
**how it works**
## Elicit helps accelerate every step of a systematic review
Step 1
### Refining the research question
Asking the right question is half the battle. Use Elicit to refine and specify your research question.
Step 2
### Gathering sources
Elicit enables you to combine multiple searches across many databases:
Use AI-based semantic search to find papers that don't overlap
Step 3
### Title & abstract screening
Elicit autosuggests screening criteria based on your research question. You can also add custom screening criteria from your protocol.
We then use AI to screen titles and abstracts, saving researchers over 80% of the time it takes to screen papers manually.
Using AI lets us be more rigorous than manual screening in some ways:
- You can see how a paper scored on each individual screening criteria, not just whether it was screened in or out overall.
- Every screening decision comes with detailed rationale and supporting quotes from the paper.
- You can refine and test criteria on a pilot set before running on all papers. This is great for scoping reviews and improving protocol design.
Step 4
### Data extraction
Data extraction is one of Elicit's superpowers. Elicit has the most accurate data extraction of any AI product.
We
... (truncated, 9 KB total)Resource ID:
41fb281a8dc4288d | Stable ID: NzhlZjYwMW