Keynote lecture - benevolent and malevolent unintended consequences of open science

Justifications for open science include: 1. It fosters transparency and reproducibility. 2. It makes access to research more equitable. 3. It increases the value of data by making it reusable. My talk will focus on another benefit that hasn't usually featured in discussions of open science, namely its role in detecting and preventing research fraud. I will present some examples where fraud was detected because data, scripts and/or peer reviews were openly available. Discussions of fraud must also, unfortunately, consider use of open data by paper millers and other fraudsters, who seize the opportunity to churn out low-quality formulaic articles. Should we make data less open to prevent such abuse, or are there other solutions?