Artificial intelligence (AI) has been called “the new electricity”—a technological invention that promises to fundamentally transform our lives and the world we live in. The resurgence of investment and enthusiasm for artificial intelligence, or the ability of machines to carry out “smart” tasks, has been driven largely by advancements in the subfield of machine learning. Machine learning algorithms can analyze large volumes of complex data to find patterns and make predictions, often exceeding the accuracy and efficiency of humans trying to accomplish the same specific task. Driven by tremendous growth in data collection and availability as well as computing power and accessibility, artificial intelligence and machine learning applications are rapidly growing in a wide range of fields, including retail (e.g., predicting consumer purchases), the automotive industry (e.g., self-driving cars), and health care (e.g., automated medical diagnoses).
These tools are beginning to be applied to environmental health in areas such as characterizing sources of pollution, predicting chemical toxicity, estimating human exposures, and identifying health outcomes. There are challenges in applying these tools to inform environmental health research and decision making. Fundamental issues of data availability, quality, bias, and uncertainty in the data used to develop machine learning algorithms are compounded by lack of transparency and interpretability of artificial intelligence systems, which can result in misleading or inaccurate results and diminish trust in artificial intelligence systems by humans who could potentially use them.