Elicit

Elicit

AI research assistant that automates literature review and evidence synthesis

Visit Product
134 upvotes 2,183 views

About

Elicit is an AI research assistant built specifically for academic and scientific research. It uses language models to help researchers find relevant papers, extract key information, and synthesize findings across large bodies of literature — dramatically reducing the time required for systematic reviews and evidence-based research.

Unlike general search engines, Elicit understands research questions semantically and can find papers that are conceptually relevant even when they don't contain exact keyword matches. For each paper found, Elicit can extract structured information such as outcomes, study population, methods, and key findings — populating a research table automatically.

Elicit is widely used by scientists, policy analysts, medical researchers, and academics who need to survey large literatures efficiently. It is particularly valuable for systematic reviews, meta-analyses, and any project that requires synthesizing evidence from dozens or hundreds of papers.

Product Features

- Semantic search across 200 million academic papers
- Automated data extraction from papers (outcomes, methods, populations)
- Research table generation from search results
- Paper summarization and key finding extraction
- Notebook feature for organizing research projects
- Citation export in BibTeX, CSV, and other formats
- Concept clarification for technical terms in papers
- Related paper suggestions based on semantic similarity
- Integration with Zotero and other reference managers

About the Publisher

Elicit is built by Ought, a nonprofit research organization focused on using machine learning to aid open-ended reasoning and support human oversight of AI. Founded in 2017 and based in San Francisco, Ought's mission is to develop AI tools that augment human reasoning rather than replace it. Elicit emerged from Ought's research into how AI can make the scientific community more productive and help evidence-based decision making scale to more of the world's problems.