25 years ago, I did a back-of-the envelope calculation during a party and came up with roughly 2040. I didn't know that Kurzweil had already predicted 2029. Onc

25 years ago, I did a back-of-the envelope calculation during a party and came up with roughly 2040. I didn't know that Kurzweil had already predicted 2029. Once I heard about it, I felt Kurzweil was over-optimistic, and stuck to 2040. Recently, though, it has started to seem that AGI before 2029 is more likely than 2040 or after.