AI evidence extraction

Generative AI in Academic Research

Angela Elkordy, Doug van Dyke, Katie Giradot, L. Anderson · Advances in computational intelligence and robotics book series · 2025

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

9/10

Relevance

1/4

Quality (LMQS)

Evidence

Citations

0.00

FWCI

This is an AI analysis. Read the peer-reviewed original at the publisher: https://doi.org/10.4018/979-8-3373-5092-9.ch004

Methodology & findings

Study design

Reflective practice, tool trials, and theoretical framework analysis

Main result

The chapter demonstrates that generative AI tools can enhance scholarly productivity across research stages. It specifically notes that "large language models (LLMs) like ChatGPT and Claude, along with specialized tools like Scite and Elicit, which streamline research from conceptualization to data analysis" can serve as research partners, while emphasizing the need for "human oversight and critical engagement to preserve researcher agency, rigor, and scholarly standards."

Reports effect sizes.

Research paradigm

Interpretivist/critical realist

Author conclusions

The authors conclude that generative AI can be effectively integrated into academic research workflows when properly managed. They advocate that researchers should use AI tools strategically, stating the work "advocates for human oversight and critical engagement to preserve researcher agency, rigor, and scholarly standards, filling a critical literature gap with structured, actionable guidance."

Risk of bias

Potential selection bias in tool trials (not systematically selected); Author reflexivity bias (reflective practice methodology); Lack of empirical validation of claims; No comparison of tool effectiveness; Potential bias from selective tool selection; Author perspective bias in evaluating AI tools; Limited empirical validation of claims

Limitations

The chapter critically addresses limitations by noting "epistemic challenges of using AI in scholarly work, including issues of accuracy, reliability, bias, and academic integrity," indicating these are significant constraints on the application of generative AI in academic research.

Open questions raised

The chapter identifies a gap in "structured, actionable guidance" for integrating generative AI into academic research while maintaining scholarly standards and addressing epistemic challenges.
The chapter addresses a "critical literature gap" regarding structured, actionable guidance for integrating generative AI tools into academic research workflows while maintaining scholarly standards and addressing epistemic challenges.
The chapter identifies a critical gap in the literature regarding structured, actionable guidance for integrating generative AI into academic research while maintaining scholarly integrity and researcher agency.

Data: not_statedCode: not_statedExtracted from: pdfAgreement 74%

Explore related topics

Responsible AI Use Human-AI Collaboration Conceptual