✨🤖 The Age of AGI: the upsides and challenges of superintelligence

'The advent of AGI could fundamentally alter the dynamics of our economic system'

OpenAI CEO Sam Altman believes we are on the cusp of an "intelligence age" that will bring unprecedented prosperity and capabilities to humanity. Getting here didn’t happen overnight, though the next leap might seem that way. After “thousands of years” of scientific and technological progress, he writes in a brief personal essay, humanity has learned to transform sand into computer chips capable of creating increasingly powerful AI. Superintelligence could emerge in just a “few thousand days.” Regardless of the timeline, its arrival seems inevitable to Altman and will be the most consequential development in all of history, in his view. 

Altman (in a riff off Arthur C. Clarke's famous quote that “any sufficiently advanced technology is indistinguishable from magic.”): “In the next couple of decades, we will be able to do things that would have seemed like magic to our grandparents."

The big takeaway here isn’t that superintelligence or artificial general intelligence could be here within a decade. That’s a pace of progress already being discussed. The forecasters at the Metaculus prediction market see “weak AGI” (a specialized AI system focused on specific cognitive and game-playing abilities) by 2027 and “strong AGI” (an advanced, general-purpose AI system capable of human-like or superhuman performance across a broader range of cognitive and physical tasks) by 2032. So AGI in about a thousand days.

The real news here is that when Altman discusses the challenge of superintelligence, there isn’t a mention of existential risk, which is something he used to seem quite concerned about. The closest he comes is a bit about “extremely high-stakes challenges.” Then again, he might just be referring to the other challenges that do get an earlier mention: unequal access to AI, labor market disruption, and the need for massive infrastructure development (primarily compute power and energy resources). His focus appears to have shifted towards managing these near-term impacts while maintaining an optimistic outlook on AI's long-term potential for “unimaginable” prosperity.

Imagining the economic world of superintelligence

Let’s dig into the potential economic challenges of AGI, which are pretty high stakes even if not existential, exactly.