[Note from Alexandros: The article below is a rewrite of a thread I wrote on Twitter here. The rewrite was mostly done by ChatGPT, with some edits from me. If you want to see Ryan and me experiment further with content transformations, you can tune into our livestream here, starting right after the publication of this article. Enjoy!]
Imagine a world on the brink of a metamorphic transformation, as profound as the caterpillar evolving into a butterfly. A world where everything is custom-made, tailored, and optimized for the individual. This is the potential we face as Generative AI, of which ChatGPT is the most prominent example, begins to reshape our understanding of technology, expertise, and the future itself.
As we stand on the precipice of this new era, it is imperative to understand the potential of this technology. I cannot say with any certainty that I know what will happen. I am barely able to understand what I am looking at. But I can paint for you one vision of the future.
Picture it as a wellspring of knowledge, capable of providing 95% of a correct solution for any problem humans can already solve. And with the ability to learn from its mistakes (for example by checking its results against a physics simulation or a programming code compiler), this technology can become an effective expert in a wide range of domains - from math and physics to engineering and biology.
With this newfound expertise, it has the power to refine and optimize our world in ways we could have never imagined. The clunky, mass-produced items of the past will be supplanted by bespoke creations that cater to our individual needs. The industrial revolution once alleviated starvation and improved the human condition, but mass production has always been a double-edged sword - a one-size-fits-all approach that leaves many wanting.
The rich have long been able to escape the confines of mass production, indulging in tailored experiences that elevate their lives. But the arrival of ChatGPT ushers in a new era where bespoke becomes accessible to all. With the integration of 3D printing, laser cutting, and electroplating technologies, the concept of mass manufacturing becomes obsolete, replaced by context-adapted, superior mass customization.
As this metamorphosis unfolds, we'll witness a seismic shift in the traditional division of labor. The need for experts to collaborate on large projects will be supplanted by the on-demand, selfless expertise of AI. This transformation strikes at the heart of the principal/agent problem, making organizations and discrete fields of expertise obsolete.
This revolution extends beyond physical objects, reaching into the realm of software. With generative AI, the need for programming libraries will disappear, replaced by ideal subroutines conjured on-demand. Legacy software can be refactored, simplified, and improved, ultimately leading to a more efficient and streamlined digital world.
As we ride the waves of optimization, the barriers between domains of knowledge will crumble. The potential for cross-domain translation will reveal new insights and spark innovation. The jargon walls between set theory and architecture, computing and accounting, and medicine and rocketry will be demolished, heralding a world where knowledge is fluid and interconnected.
This metamorphosis, while breathtaking, also brings with it the fear of the unknown. The journey ahead promises a civilization with far less needless complexity, but the path is fraught with uncertainty. The optimistic scenario is akin to the caterpillar's transformation into a butterfly, with the potential for greatness lying dormant within us. The pessimistic scenario has been… well covered by others.
And as we contemplate the impact of AI on science and engineering, we must also ponder its influence on the arts, politics, and international affairs. The truth is, I have no idea where this train will go.
As we cross the event horizon into the age of bespoke, we must brace ourselves for the unknown. The singularity may already be upon us, defined as the point in time when we can no longer predict the future, and what comes next is anyone’s guess.
Feel free to link me to other analyses on the future of this technology in the comments. I confess to being very uncertain about what comes next, and my highest priority right now is to better understand this alien life form we seem to be in contact with, for lack of a better term.
This article reminds me of the articles published in the newspapers under the former Eastern European communist regimes. "Wellspring of knowledge" ? Lol. Where is the "fountain of wisdom"?
Imagine, if you will, an AI employed to control indoor climate in people's homes. This AI might learn that lower oxygen levels "help" with sleep. Then imagine finding out a decade afterwards, if ever, that nonzero unusual overnight deaths (within statistical noise) were not random accidents.
This is the future of AI, and the stupid humans who will rely on it.