Comment by moregrist

> It doesn't take a tech blogger to draw a trend line through the advancements of the past 2.5 years and see where we're headed.

People did this with airplanes in the 60s, and based on that trajectory we should be exploring the outer edges of our solar system by now. Turns out the market for supersonic jets was unsustainably small and the cost/risk of space exploration is still very high.

Every sigmoidal curve looks exponential as it starts to enter the linear regime. But eventually the curve turns over, either due to limits in the technology, the marginal cost of the technology, or no clear way to further commercialize it.

I don't know that we've reached that point with AI, but a do know that extrapolating from a trend line is fraught with peril.