Future of the AI and Humanity

Two Diverging Futures for AI

Last month, two major publications were released discussing the future of our species, our role on this planet, and our relationship with AI. One is the now-famous AI 2027 paper by Kokotajlo et al., which explores the uncontrollable acceleration expected in the coming years due to breakthroughs in AI-driven coding. This, they argue, could lead to the emergence of superhuman AI researchers who significantly speed up progress—potentially taking us to an unsettling future¹.

In contrast, Narayanan and Kapoor offer a different perspective in their essay titled AI as Normal Technology. They argue that AI, like any other technology, will require time to diffuse throughout society. Despite all the hype, they suggest that no matter how revolutionary AI may seem, its adoption is ultimately constrained by societal adaptation².

I plan to use these two detailed texts as references as I continue to work through them and form my own opinions. I will update this draft along the way.

These works envision two very different worlds.

References