This month, David Corliss shares how the Data4Good practitioners are leading the way in developing ethics best practices for artificial intelligence.

Gene Pennello, a statistician, and Frank Samuelson, a physicist, look at the complexity of data-driven medical devices that employ AI algorithms—and the challenges this presents.

Rising high-school senior Jaya Kolluri shares how she thinks AI tools such as ChatGPT can complement traditional teaching methods.

In the first half of 2023, the machine learning programs ChatGPT and GPT-4 changed the landscape of artificial intelligence research seemingly overnight. Judea Pearl’s research bridges the subjects of statistics and artificial intelligence and highlights the importance of causality in both settings. Dana Mackenzie, Pearl’s co-author for The Book of Why, interviews him here to get his take on recent developments. When they wrote their book in 2018, Pearl contended machine learning had not yet moved past the first rung of the “ladder of causation.” Computers could not correctly answer queries about interventions and still less about counterfactual scenarios. Has his assessment changed?

Nathen Huang looks at the potential use of AI in the federal government—how it can help but also its possible limitations.

AI has required substantial statistical input along its entire developmental path. Here is an overview of major statistical concepts and methods integral to AI and the statisticians involved in their development.

Michael Hansen has taught AP Statistics for 25 years and high-school computer science for about 13 years. This summer, he used ChatGPT to help him write a Python application of about 3,000 lines. Here are his top 10 takeaways from that experience.

David Banks and Yen-Chun Liu examine the future of statistics in artificial intelligence for autonomous vehicles.

Ethical issues in statistical practice in the AI era require continuous attention, reflection, and dialogue among all parties involved. The ASA’s Ethical Guidelines for Statistical Practice and members of the ASA Committee on Professional Ethics are here to help.

David Corliss examines how shifting the focus to testing for and mitigating bias at the design stage instead of after code is released can help prevent many of the problems seen in machine learning and artificial intelligence.