Decoding AI Through Cybernetics

Yesterday I had the privilege of experiencing the first program in ANU School of Cybernetics new suite of one-day learning experiences, Decoding AI through Cybernetics.

Offering so much more than technical understanding of AI systems and led by Distinguished Professor Genevieve Bell , A/Professor Matthew Holt , Dr Safiya Okai-Ugbaje, PhD and Dr Gabriela Ferraro , we explored how AI systems have evolved and how they interact with wider cultural, social, economic, political and ecological systems.

It was fascinating to learn about the Dartmouth Summer Research Project on Artificial Intelligence held in 1956. This was the event that kicked off AI as a research discipline.

It was organised by John McCarthy, Marvin Minsky, Claude Shannon, and Nathaniel Rochester and brought together a few dozen of the leading thinkers in AI, computer science and information theory to map out future paths for investigation.

Among questions that emerged were materialistic visions of the human mind and attempts to highlight the peculiar and unavoidable traits of man compared to any machine.

Their proposal to the Rockefeller Foundation for a ten-week research project stated that: “The study is to proceed on the basis of the conjecture that every aspect of learning or any other feature of intelligence can in principle be so precisely described that a machine can be made to simulate it.”

What is so interesting is that a decade earlier, some of the same players at Dartmouth had also participated in the Macy Conferences.

This was a very different set of conversations about the future curated by anthropologists Margaret Mead and Gregory Bateson, in collaboration with Norbert Wiener who coined the term cybernetics.

A key feature of the cybernetics approach to imagining the future was the diversity of the participants.

There were mathematicians and philosophers, physicists and psychologists, anthropologists and historians, hailing from North America including Mexico, Europe and Asia. Participants were at different points in their careers, had different lived experiences and considered futures built on the intersections between fields and existing ideas.

Unlike Dartmouth, whose participants were dominated by computer scientists that wanted to make a machine that can think like a human, earlier conversations were characterised by a systems approach.

At the Macys Conferences, participants discussed how society should integrate and develop artificial intelligence, with a focus on how humanity might be enhanced. Economic benefit was a secondary consideration.

This is so different to how AI applications are positioned and employed in the majority cases today, where profit, speed and efficiency often trump societal benefits.

Another difference is that there were fewer voices, and fewer questions at Dartmouth.

As we discussed yesterday, many ideas were lost in the transition from cybernetics to artificial intelligence, most notably the idea of dynamic systems that included technology, people and ecology; the idea of feedback; and the idea that the world you were making should be subject to critical inquiry.

At the School of Cybernetics, we are reimagining cybernetics as an important framework for navigating complex societal transformations safely, sustainably and responsibly.

By decoding AI through cybernetics, this course offers a way to explore the intersections of the technological, the ecological and the human and to consider today’s rapidly changing technological systems in relation to so many other systems they interact with.

It’s a thought-provoking and highly stimulating experience and one I would highly recommend to anyone seeking a deeper understanding, working out how to find a new way through the uncertainty generated by novel technologies and engaging with AI from fresh perspectives.

Discover what role you can play in building the better, fairer and safer systems of the future!

Learn more about the School of Cybernetics’ learning experience series, Navigating Cybernetic Futures, and what's on the program for September and October at https://cybernetics.anu.edu.au/education/microlearning/