12,445 papers · continuously updated · last export: 10 Aug 2026livingmeta.ai
← Browse all papers
AI evidence extraction

AI in Research: Smarter Strategies for Literature Reviews

Bronte Chiang, Kathleen James · University of Calgary · 2026

AI-generated evidence extraction, verified across multiple analytical personas. Not a substitute for the peer-reviewed original.

9/10
Relevance
0/4
Quality (LMQS)
I
Evidence
0
Citations

This is an AI analysis. Read the peer-reviewed original at the publisher: https://doi.org/10.11575/prism/51024

Methodology & findings

Study design

Workshop-based instruction with learning objectives designed to develop participant competencies in evaluating and integrating AI tools into literature review processes.

Primary method

instructional_design

Main result

The workshop demonstrates that AI tools including "ChatGPT, Scite, and ScopusAI can assist with topic exploration, research question refinement, identifying key themes, summarizing sources, and generating synthesis." However, critical limitations exist: "hallucination risks, lack of access to paywalled content, and outdated training data" must be addressed through careful evaluation and comparison with academic databases.

Research paradigm

pragmatist

Author conclusions

The authors conclude that participants should "Develop strategies for integrating AI responsibly into citation management, synthesis, and argumentation while maintaining critical engagement with scholarly sources." This reflects the core message that "AI can help with organizing citations and structuring arguments, while underscoring that critical thinking and scholarly judgment remain essential."

Limitations

  • The workshop highlights significant limitations of AI tools, including "hallucination risks, lack of access to paywalled content, and outdated training data." Participants are taught to evaluate "the affordances and limitations of AI tools in relation to the literature review process, including issues of hallucination, access restrictions, and outdated data."

Open questions raised

  • The paper identifies the need for strategies to evaluate AI-generated outputs against academic databases and to develop responsible integration approaches that maintain scholarly rigor despite AI limitations.
Data: not_statedCode: not_statedExtracted from: pdfAgreement 72%

Explore related topics

Systematic Review AutomationLiterature DiscoveryDesign Science

Related papers

← Browse all papers