The advancements in AI tend to happen in intermittent large leaps, but people tend to make linear projections based on the most recent leap. That’s why there’s

The advancements in AI tend to happen in intermittent large leaps, but people tend to make linear projections based on the most recent leap. That’s why there’s always a big hype cycle followed by disillusionment and some sort of AI winter. I personally think 2100 is extremely pessimistic, but all those projections of full self-driving by 2025 were certainly even more extreme in their optimism.