Rupert Read

"What is to be done?: Looking to a relocalised future, in an emerging age of AI- and climate- extremes"

It is very doubtful, on both philosophical grounds, and broadly ecological grounds, that we will create a genuine machine superintelligence. As yet, we have nothing even worth calling artificial intelligence. AIs don't even have as much intrinsic, self-generated purpose as an amoeba. And yet: AI obviously has fantastic and accelerating digital power. It is a likely accelerator of most trends toward the polycrisis (and toward the rupturing of planetary boundaries)-including disinformation, polarisation, the likelihood of accidental nuclear war, the possibility of complete financial/economic meltdown, dangerous synthetic biology, high energy use, and more. Furthermore, while the danger of AI consciously driving such crises is over-hyped, the danger of those who profit and gain power from AI, either consciously or unconsciously, is probably under-hyped. This is a neglected respect in which AI is, in practice, a self-accelerating catalyst of polycrisis trends.

Precisely because AI is not actually "intelligent," it is immensely dangerous. Precisely because AI has, as yet, no actual purchase on the physical world/reality that it 'hallucinates' (and there is no prospect of this changing until the AI-research-frontier drops its obsession with LLMs), 'Stupid AI', which is intrinsically not 'aligned', could quite easily destroy or at least massively diminish humanity in the various ways noted above: including through a reign of deep-fake terror, which may be all the more insidious by way of being personalised, but also perhaps in less obvious and slower, yet potentially profoundly devastating ways, such as through an attritional impact, at scale, on the quality/felt-reality of human relationships.

This talk, coming from a background in the philosophy of Wittgenstein, the thought-leadership of Daniel Schmachtenberger and of my collaborator Emile P. Torres, and the action-orientation of the Climate Majority Project, explores the insanity of a world in which AI is largely unregulated. It also explores the consequences of this in terms of our action-pathways: towards policy-action and political contestation, as well as towards 'transformative/strategic adaptation'-i.e. towards looking to the civilisation which can emerge out of the pupa (or, failing that, out of the ashes) of this one.

See rupertread.net/writings/2023/what-is-the-climate-majority-project/ if you want to understand the CMP, that I now head up.

Bio

Rupert Read is Co-Director of the Climate Majority Project and Emeritus Professor of Philosophy at the University of East Anglia. He is the author of several books, including This Civilisation is Finished, Parents for a Future, Why Climate Breakdown Matters and Do you want to know the truth? The surprising rewards of climate honesty.