Elicit

Picture the worst week of a systematic review: a few thousand abstracts to screen, then a spreadsheet of extracted outcomes to build by hand. That's the workflow Elicit is clearly built around. Its Pro plan advertises a "Dedicated Systematic Review Workflow that can screen 5,000 papers," and the company says it automates screening and data extraction while only "partially supporting search and report generation." That honesty is useful. You're not handing over the whole review. You're handing over the parts that eat the most hours, and Elicit says researchers report up to 80% time savings on them.

Around that core sits a broader research tool. You can run semantic search across more than 138 million academic papers and 545,000 clinical trials, so you don't need the perfect keyword string to get relevant hits. The Research Agent pulls evidence from scientific databases, journals, public data and your own uploaded documents, then produces cited reports, tables, figures, calculations and slides. Research Reports follow a process "inspired by systematic reviews," and you can change which papers and which information they cover. Every AI-generated claim gets sentence-level citations, which matters a lot if you have to defend a finding to a reviewer or a regulator. Elicit says it can find up to 1,000 relevant papers and analyze up to 20,000 data points at once.

The free Basic tier is genuinely usable: unlimited search, summaries and chat with full-text papers, plus Zotero import. The catch is that the heavy features, the Research Agent and Reports, are rationed on free and scale by usage multiplier on paid plans, and things like Routines, custom extractions and API access only show up at Pro ($39/user/mo billed annually). If you do evidence synthesis for a living, Elicit is one of the more focused tools you can try. If you just want quick answers, it's more machinery than you need.