Imagine if we spent all our time worrying about whether a machine could write poetry like Shakespeare, while completely missing the fact that it was already rewriting the rules of global warfare or the entire world economy. In this episode, we dive into a landmark paper that argues our obsession with 'human-level intelligence' might be a massive category error. Instead of asking how smart AI is, we should be asking how much it is about to change the way we live [1, 2].

Ross Gruetzemacher and Jess Whittlestone propose a new framework to help us make sense of this shift. They move away from vague terms like 'Artificial General Intelligence' and introduce 'Transformative AI' (TAI), defined not by cognitive ability but by its power to create practically irreversible change in societal trajectories [3, 4]. By drawing on historical parallels like the industrial revolution and the invention of nuclear weapons, the authors provide a clear taxonomy for the different ways AI might disrupt our future [5, 6].

This framework isn't just academic—it is a tool for policymakers and researchers to prioritize the risks and opportunities that matter most before they become 'locked-in' to our social fabric [11, 12]. Tune in as we unpack how to distinguish between hype and history-making technology.

Ref:
Ross Gruetzemacher, Jess Whittlestone. The transformative potential of artificial intelligence. Futures, 135, 2022, 102884. ISSN 0016-3287. https://doi.org/10.1016/j.futures.2021.102884