Elicit Review

A research assistant for the literature review, not the chat box

Elicit is not a general-purpose chatbot — it is a specialist built for one of the most tedious jobs in research: the systematic literature review. Point it at a question and it searches roughly 138 million papers, screens them against criteria you set, and pulls the relevant findings into a structured table where every cell is backed by a sentence-level citation. For UK academics, policy analysts and evidence teams, that turns days of manual screening into an afternoon.

Strengths

The sentence-level citations are the standout: Elicit shows you exactly where each extracted claim came from, so the output is defensible rather than a black-box summary. The paper database is vast, the screening workflow mirrors how real systematic reviews are run, and results export cleanly into reference managers. On security it is better documented than most research tools, with a clean SOC 2 Type II report and a stated policy of not training on your uploads.

Limitations

Elicit's focus is also its constraint. It is built for academic evidence synthesis, so it is the wrong tool for general writing, current-events research or anything outside the published literature. The professional tiers are noticeably more expensive than a £16–20 general AI subscription, which makes sense for funded research teams but stings for individuals. And as with any AI summary, you should still read the source papers before you rely on a finding.

UK data and GDPR

Elicit is more transparent than many research tools: it holds a clean SOC 2 Type II audit and says uploaded content is not used to train models, with third-party model providers held to zero-retention terms. The caveat for UK users is residency — data is processed in the US and there is no published EU/UK data-residency option, so for personal or confidential material you should review the current DPA and security documentation, and avoid uploading anything you are not licensed to share.