Related Experiment Video
Updated: Aug 5, 2026
08:58
Artificial Intelligence Approaches to Assessing Primary Cilia
Published on: May 1, 2021
Twelve quick tips for AI-assisted coding in science
Eric W Bridgeford1, Iain Declan Campbell2, Zijiao Chen1
1Department of Psychology, Stanford University, Stanford, California, United States of America.
Plos Computational Biology
|July 27, 2026
Summary
This paper offers twelve practical tips for AI-assisted coding in scientific research. These guidelines ensure AI accelerates software development while maintaining scientific validity and code quality.
Area of Science:
- Scientific Computing
- Software Development
- Artificial Intelligence
Background:
- AI coding tools can accelerate software development.
- Their application in scientific computing raises concerns about code quality and scientific validity.
Purpose of the Study:
- To provide twelve practical tips for AI-assisted coding in scientific research.
- To balance AI capabilities with scientific and methodological rigor.
Main Methods:
- The paper outlines strategies across the development cycle: problem understanding, context management, testing and validation, and code quality assurance.
- Emphasis is placed on human agency, robust validation, and domain expertise.
Main Results:
- Twelve practical tips are presented to guide researchers in using AI for coding.
- These tips cover strategic leveraging of AI throughout the software development lifecycle.
Conclusions:
- Researchers can harness AI for faster software development.
- Adhering to these principles ensures code reliability, reproducibility, and scientific validity, upholding research integrity.
