Mode
Text Size
Log in / Sign up

Elicit AI tool shows promise for data extraction but lacks precision for independent systematic reviewsAI Tool Shows Promise for Speeding Up Medical Research

AI-generated summary of the cited source, checked by automated accuracy review. How we work

Key Takeaway
Note that Elicit AI shows promise for data extraction but is not a standalone replacement for systematic reviews.

This systematic review evaluates the performance of the Elicit AI tool in replicating systematic review processes. The evaluation focused on search overlap, screening accuracy, and data extraction capabilities compared to a recently published systematic review by the same research group.

Findings indicate that Elicit produced poor search overlap, with only 158 of 5285 studies found. While the tool showed moderate sensitivity and accuracy during abstract and full-text screening, it exhibited poor precision with a skew toward false positives. However, the data extraction tool performed well, providing helpful information for 85% of values even without exact matches. The process using the tool took 3.16 hours to complete the full systematic review.

The authors note that Elicit is not ready as a standalone tool to replace the systematic review process. While it shows promise as an adjunct tool for screening and extraction, its limitations in precision and search overlap mean it requires human oversight. Clinical researchers may consider it as a supportive tool rather than a primary replacement for standard systematic review methodologies.

Researchers evaluated a new artificial intelligence tool called Elicit to see how well it could perform the tasks of a systematic review. This type of review is a rigorous process where experts gather and summarize all available evidence on a specific medical topic. The study looked at how well the tool could find studies, screen them, and extract data compared to traditional methods.

The results showed that while Elicit was not great at finding the exact same studies as humans, it was helpful for extracting data. It provided useful information for 85% of the values it checked. The tool also significantly reduced the time needed to complete the review process, cutting it down to about 3 hours. However, the tool had a tendency to include too many irrelevant results during the screening phase.

Because the tool still produces many false positives, it is not ready to replace the human-led review process. Instead, it is currently best used as an extra tool to help researchers work more efficiently. It can assist with screening and data collection, but human oversight remains necessary to ensure accuracy in medical research.

What this means for you:
Elicit can speed up data extraction and research tasks but requires human oversight to ensure accuracy.

Common questions

Can this AI tool replace human researchers?

No, the tool is not ready to replace the systematic review process on its own. While it can help with screening and data extraction, it still produces many false positives. It is currently best used as an extra tool to help researchers work more efficiently rather than replacing human oversight.

How much time can the AI tool save?

The study found that the tool could help complete the full systematic review process in 3.16 hours. This shows it has the potential to save significant time for researchers during the data collection and extraction phases of their work.

How accurate is the data extraction with this tool?

The data extraction tool showed promising results. It provided helpful information for 85% of the values it checked, even in cases where it did not find an exact match. This makes it a useful tool for helping researchers gather information more quickly.

Study Details

Study typeSystematic review
EvidenceLevel 1
PublishedOct 2026
View Original Abstract ↓
Background: Systematic reviews are critical for informing evidence-based medicine, policy, and future research. However, conducting a systematic review is labour- and cost-intensive and error-prone. Elicit is an AI tool that claims to replicate the systematic review process. Methods: Our study investigated Elicit's ability to replicate a systematic review by comparing its results with those of an existing systematic review recently published by our research group. We calculated accuracy, precision, sensitivity, error rate, and person-time savings at each stage of the systematic review. Results: At search, Elicit had poor overlap, finding 158 of the 5285 studies. For abstract and full-text screening, Elicit had moderate sensitivity and accuracy but poor precision, with a skew toward false positives. Elicit's data extraction tool was promising, with sentence-specific citations, and provided helpful information for 85% of values even when it did not find exact matches. Elicit offered person-time savings, taking 3.16 hours to complete the full systematic review process. Conclusions: Elicit shows promise as an adjunct screening and extraction AI tool, but in its current form, it is not ready as a standalone tool to replace the systematic review process.
Free Newsletter

Clinical research that matters. Delivered to your inbox.

Join thousands of clinicians and researchers. No spam, unsubscribe anytime.