Sam Altman - “once we’ve built this general intelligence, we will just ask it how to generate an investment return”

A recent comment by OpenAI CEO Sam Altman has ignited a fierce debate about the practical limits of artificial general intelligence (AGI). In a viral statement, Altman suggested that "once we've built this general intelligence, we will just ask it how to generate an investment return." This vision positions AGI as a potential oracle for solving high-stakes, complex problems like financial forecasting and capital allocation, implying it could identify opportunities invisible to human experts.

However, this claim has met with significant skepticism from professionals and observers. The core criticism, as articulated in a popular Reddit discussion, questions the fundamental premise: if a problem currently has no reliable solution from a consensus of human experts, how could an AGI, whose intelligence is built from human knowledge, magically find one? Critics argue that in domains like finance, an AI model would essentially be synthesizing and extrapolating from existing human data and strategies. It would not possess some inherent, supernatural insight beyond the collective wisdom of hundreds of career financiers, quants, and economists whose work forms its training corpus.

The debate highlights a critical fork in the road for AI expectations. Proponents of 'transformative AGI' believe sufficiently advanced systems will exhibit emergent capabilities and reasoning that transcend their training data, potentially revolutionizing fields like scientific discovery and complex system optimization. Skeptics counter that AI, even at a general intelligence level, will remain a powerful tool for augmentation and acceleration—excellent at enhancing human knowledge and speeding up processes—but not a replacement for novel, domain-specific human expertise in ill-defined or unprecedented problem spaces. Altman's quote has become a litmus test for which of these visions one subscribes to.