This column is written for those interested in learning about the world of Data for Good, where statistical analysis is dedicated to good causes that benefit our lives, our communities, and our world. If you would like to know more or have ideas for articles, contact David Corliss.

David Corliss is the principal data scientist at Grafham Analytics. He serves on the steering committee for the Conference on Statistical Practice and is the founder of Peace-Work.

Ask any bot—ethical AI is a big deal.

The development and use of AI arguably goes back to the 1940s and code-breaking devices. Recently, as powerful new tools such as large language models have gained public visibility, there has been tremendous growth in the concern about AI and its capabilities, uses, and abuses. As a result, ethical AI teams are becoming firmly established in academia, industry, and government. With more AI applications being released every day, #DataForGood practitioners in every area are leading the way in developing and promoting ethical best practices.

A great review article from the producers of the Open Data Science Conference, titled “The AI Ethics Boom: 150 Ethical AI Startups and Industry Trends,” provides a plethora of information about and analysis of industry trends. The article highlights the Ethical AI Database, listing more than 200 ethical AI start-ups in both commercial and NFT spaces.

A number of independent not-for-profit organizations were first responders to growing concerns about the ethics and effects of artificial intelligence. Distinguished ethical AI leader Timnit Gebru has been a driving force in this space, founding the Distributed Artificial Intelligence Research Institute for community-centered AI research independent of the industry. She also founded Black in AI for Black professionals in artificial intelligence.

The number of organizations continues to grow rapidly. The post-pandemic technology landscape makes it easy to get involved with a group working in your particular area of interest, no matter where you and the ethical AI organization are located.

Major research universities have been fertile ground for ethical AI research. An early leader in this space is Stanford’s Institute for Human-Centered AI (HAI). Established in 2019, HAI supports research, educational classes and materials, and policy development. A focus at Stanford is the effect of AI on the human experience and society as a whole. HAI is active in hosting events, including webinars and a boot camp for congressional advocacy.

Oxford University’s Institute for Ethics in AI works globally to connect ethicists, philosophers, and social scientists with academia, industry, and government. Influenced by the impact of philosophers on the establishment of ethical practices in medicine decades ago, the Oxford Institute fosters the establishment of ethics in AI as a standardized area of practice. They seek to have ethical training included as an integral part of tech training and are developing a program for visiting professors to participate in.

Developing partnerships with academia and industry is a focus of the University of Michigan Institute for Data Science. As a leading public research university, its mission includes a strong commitment to extending beyond academic research to serve intersectionality across a wide range of interests to the public. The institute recently hosted a one-day forum titled “From Theory to Practice: Building Ethical and Trustworthy AI,” that could serve as a model for others. It featured presentations from Rocket Company’s dedicated ethical AI team.

It’s been my observation that industry tends toward creating a team within their IT organization while most university and government programs seek to be interdisciplinary, bringing together researchers from many departments. In response to this need, the University of Michigan Institute for Data Science is developing training materials that will help companies set up their own ethical AI teams.

I asked ChatGPT if ethical AI is important. Its response was a page and a half long, and I didn’t read it before finishing this column to avoid any influence on my writing. Given how LLMs work, the long, detailed response shows a lot of work is being done in this area. ChatGPT has a list of limitations on its home page, including “May occasionally generate incorrect information. May occasionally produce harmful instructions or biased content. Limited knowledge of world and events after 2021.” This shows the impact ethical AI is having on the tools and how they are presented to the public, reflecting concerns about how the results might be used.

As the use of AI grows—along with increasing public awareness of the ways it affects our lives every day—ethical AI will become even more important. There are myriad ways Data for Good practitioners can get involved, including while doing independent research or working for a university, not-for-profit, or ethical AI team at a tech-driven company. Whatever your area of work, ethical AI is now part of it as an essential element of Data for Good.