This short story about “AI on blockchain” was originally published at my Substack on November 24, 2021, and is republished here with very minor changes as the first entry on my Mirror blog.
In the half a year since I wrote it, the concerns implicitly raised in it have become far more acute. Predicted timelines for the appearance of “weak AGI” at Metaculus have compressed sharply in the past few months, as the scaling hypothesis - that is, the concept that banal increases in model size lead to discontinuous increases in capability - has failed to break down. Meanwhile, there are now projects aiming to implement fully on-chain AI. One such project is Giza, which wants to build them on top of a second/third layer on Ethereum that allows for intensive computations while inheriting Ethereum’s security and decentralization.
“Why is putting artificial intelligence on chain a good idea?” asks one piece on this topic. Why not indeed. Think of the possibilities! ;)