Frontiers in Artificial Intelligence, Journal Year: 2025, Volume and Issue: 8

Published: May 12, 2025

The exponential growth of scientific literature presents challenges for pharmaceutical, biotechnological, and Medtech industries, particularly in regulatory documentation, clinical research, systematic reviews. Ensuring accurate data extraction, synthesis, compliance with industry standards require AI tools that not only streamline workflows but also uphold rigor. This study evaluates the performance designed bibliographic review, assessing their impact on decision-making, compliance, research productivity. assessed include general-purpose models like ChatGPT specialized solutions such as ELISE (Elevated LIfe SciencEs), SciSpace/Typeset, Humata, Epsilon. evaluation is based three main criteria: Extraction, Comprehension, Analysis Compliance Traceability (ECACT) additional dimensions. Human experts established reference benchmarks, while Evaluator ensure objective measurement. introduces ECACT score, a structured metric reliability analysis, reporting documentation. Results demonstrate consistently outperforms other tools, excelling precise deep contextual comprehension, advanced content analysis. ELISE's ability to generate traceable, well-reasoned insights makes it well-suited high-stakes applications affairs, trials, medical where accuracy, transparency, are paramount. Unlike provides expert-level reasoning explainability, ensuring AI-generated align best practices. efficient retrieval lacks precision complex limiting its use decision-making. Epsilon, SciSpace/Typeset exhibit moderate performance, variability affecting critical applications. In conclusion, enhance writing, interpretation, human oversight remains essential validate outputs standards. For integration must strike balance between automation expert supervision maintain integrity, adherence.

Language: Английский