I think a useful way of thinking about this might be looking back to self driving cars. In the mid-2010s we had a boom in investment in self-driving cars. It ha

I think a useful way of thinking about this might be looking back to self driving cars. In the mid-2010s we had a boom in investment in self-driving cars. It happened basically because the technology had mostly got there, people could see how it might eventually be solved and so there was a rush in. Some people made bold and self-serving claims about how quickly it would be solved. But in reality it wasn't solved. No car could self drive for any reasonable amount of time. There was a step forward - decent cruise control was suddenly way better and as long as you viewed it as that it was a great product. It wasn't self driving. A decade later, we kind of have self-driving. In some situations, in some locations we do actually have Waymo taxis. But (a) the guys who claimed they could do self-driving back in the mid 2010s (Tesla) never actually got there (despite claiming they did), and a lot of investment was made on the basis of the pay-offs of self-driving where what was really delivered was just good cruise control. For me, AI is the same. We've got some really neat technology that is going to enhance our productivity right now. If you embrace it that way and adopt that as a strategy I think you'll see massive payoffs right now. If you think you're getting full AI employees, and you invest in that way, you're going to be disappointed. Worth noting of course a lot of these CEOs are talking like they think the full AI employee is here simply because they're trying to pump their stock and they're behaving very differently to how they're speaking.