GAO Suggests Ways to Embrace Machine Learning in Drug Development

The Government Accountability Office (GAO) laid out policy options for dealing with the challenges of machine learning in drug development— noting that uncertainty about potential regulation of machine learning is hindering its adoption.

One option would be to simply allow drug companies to continue their efforts in machine learning without any intervention, according to the report prepared along with the National Academy of Medicine.

But this could exacerbate current problems with machine learning, such as bias and underrepresentation of certain groups in existing data, GAO said.

The joint report called on policymakers to develop a “clear and consistent message” for regulating machine learning in drug development and for collaboration with stakeholders to create uniform standards for data and algorithms.