Sustainability and AI - public conversation about "Dark Machines" with author Victor Galaz

 .

This week I moderated a talk about a just-out book on sustainability and AI, "Dark Machines: How Artificial Intelligence, Digitalization and Automation is Changing our Living Planet". This activity is part of the KTH-housed cross-disciplinary research centre Digital Futures' reoccurring book conversations where authors are invited to present and talk about their books. Dark Machines is published by Routledge (which unfortunately means that it's quite expensive) and the book's own homepage can be found here.

The author, Victor Galaz is a political scientist and a climate scientist at the Stockholm Resilience Center. What I appreciate about the book is that it's a sustainability-first book about AI, and not an AI-first Pollyannish book about how AI can solve all environmental problems (past, present and future*). Galaz' instead believes that calls such as "AI for Good", "AI for Climate" and "AI for the Planet" mostly represents a very shallow views of what AI could do (perhaps, maybe, if we all just came together) and of the likely effects of AI in relation to sustainability (e.g. bordering on greenwashing). Galaz' book instead starts off with the state of the planet (not good), the direction we are going in (it's getting worse year-on-year), what AI already is used for today, the impacts on the living planet (usually quite or very problematic) and what AI could do and sometimes actually does (hopeful - but not enough). 

The book runs through different aspects of AI (and digitalisation and automation), but the definition of "AI" is broad and fuzzy and I would personally not refer to it as "AI" but rather just as "computing (including AI)". So computing (now-with-more-AI) changes agriculture and land-use, how we monitor and measure this-and-that (in the often misguided hope that things will change for the better just because we have more precise data), algorithmic bias, cryptocurrencies and blockchain (""blockchainification" of carbon offsets [is] the combination of two bad ideas, merged into one terrible [idea]"), how capital and financial markets shape the planet, how (unintentionally inaccurate) misinformation and (deliberately false or misleading) disinformation increases, how data about emotions are collected and mined and how polarisation in society increases. And how we use computing to seek increased control, optimisation, efficiency, and productivity, how black box models makes just about everything less transparent and harder to understand (decreasing the chances of (democratic) control) and how rewards and risks are redistributed (leading to increased inequality). And much more that is troubling and problematic. 

The book has ten chapters and the second to last chapter discusses resistance while the last chapter outlines what "planetary responsible AI" could be. The remainder of the book is a litania of problems and troubles. The discussion we had at the event was for the most part gloomy because while there are some really nice examples of AI being used for beneficial purposes, there are many more examples of AI being used to further current (unsustainable) goals in society. 

We organised the 10th International Conference on Information and Communication Technologies for Sustainability (ICT4S) in Stockholm in June last year. I reached out to Victor at the time to invite him as a keynote speaker with the intention that he would talk about his (what I thought was his) just-to-be-published book. The conference unfortunately didn't fit his schedule, so it was very nice to (finally) have the opportunity not just to hear him talk about the book but to also read it and have the privilege to also talk to and query the author. 

Here's the blurb from the book's back cover:

This book offers a critical primer on how Artificial Intelligence and digitalization are shaping our planet and the risks posed to society and environmental sustainability.

As the pressure of human activities accelerates on Earth, so too does the hope that digital and artificially intelligent technologies will be able to help us deal with dangerous climate and environmental change. Technology giants, international think-tanks and policy-makers are increasingly keen to advance agendas that contribute to “AI for Good” or “AI for the Planet." Dark Machines explores why it is naïve and dangerous to assume converging forces of a growing climate crisis and technological change will act synergistically to the benefit of people and the planet. It explores why AI and associated digital technologies may lead to accelerated discrimination, automated inequality, and augmented diffusion of misinformation, while simultaneously amplifying risks for people and the planet. We face a profound challenge. We can either allow AI accelerate the loss of resilience of people and our planet, or we can decide to act forcefully in ways that redirects its destructive direction.

This urgent book will be of interest to students and researchers with an interest in Artificial Intelligence, digitalization and automation, social and political dimensions of science and technology, and sustainability sciences.

* That's obviously a joke, because however powerful AI is, it could hardly solve any problems in the past...

.