A Just Transition: What does it mean for AI and Evaluation?

I had the pleasure of speaking on “A Just Transition: What does it mean for AI and Evaluation” at the European Evaluation Society’s Autumn Online Event last week. The overall conference theme centered on the role of Evaluation in A Just Transition.

What is the “Just Transition” framework?

“Just Transition is a vision-led, unifying and place-based set of principles, processes, and practices that build economic and political power to shift from an extractive economy to a regenerative economy. This means approaching production and consumption cycles holistically and waste-free. The transition itself must be just and equitable; redressing past harms and creating new relationships of power for the future through reparations. If the process of transition is not just, the outcome will never be. Just Transition describes both where we are going and how we get there.” – Climate Justice Alliance

What does AI have to do with A Just Transition?

It’s important to not see AI as just another tech tool, because it has much bigger implications in terms of equity and inclusion, privacy, jobs, the environment, media and information integrity, and our overall societies. AI is everywhere, making it hard to pin down and wrangle. We need to unpack current AI narratives and the world views behind them if we are to understand AI trends in terms of where they come from, where they are going, and how they will affect individuals and societies now and in the future.

AI has implications for the profession of evaluation, both for how we use AI and for how it will alter our profession, the tasks we do, and the playing field. We can use AI to our advantage in many ways, but we must be cognizant of the potential challenges and harms. There is a great deal of interest in AI in the field of evaluation, yet we are still mostly playing around at the edges. We should be exploring and unpacking AI at a systems level.

AI is an extractive economy. In addition to drawing on huge quantities of energy and water to run massive data centers, AI systems require the extraction of data from our bodies, our minds, our purchases, our art, our statements, our actions, and our locations to fuel their outputs. Evaluators could use a “Just Transition” (or similar) framework to draw parallels and explore what the transition to an AI economy might look if done in a just and equitable way, with greater participation and accountability and respect for people and planet.

Evaluation is currently behind in its understanding of this phenomenon. Evaluators and evaluation societies should deepen their understanding of AI and its current and future implications. We have a role in contributing to a more just transition to the AI present and future.

Click on the image below to see the slides and accompanying notes. (ChatGPT/DALL-E generated some pretty amazing artwork for the slides!) If you joined the EES Autumn convening, you can access the recording here. I’d love for us to explore some of these ideas more at future gatherings.

Thanks to Lina Srivastava, Zach Tilton and Nancy MacPherson for their review and suggestions on this talk and the accompanying slides.