I follow his work and subscribe to his newsletter. His podcast with Toni is pretty good too. In that work you see him form his overall thesis that culminates i

I follow his work and subscribe to his newsletter. His podcast with Toni is pretty good too. In that work you see him form his overall thesis that culminates in these slides. It would be interesting to hear the talk behind these slides and any audience questions. The big takeaway I got from this is that tech is still eating the world and there is still a lot of potential. It is not necessarily crypto (that's at least 5 years out) or necessarily the metaverse (probably 5 years out or more). Meta has invested huge amounts of money into VR when compared to the investment Apple made in the original iPhone. But Internet enabled commerce has a huge margin to capture based on total retail spending. There is an unbundling of products from aggregators and a too many choices problem probably. A big question is how do you find customers which is basically the "ad" market and the first party data problem. These markets have huge margins but you can leverage the gatekeepers to build successful billion dollar businesses. Another sector is the continued digital transformation of current businesses which has a large cloud workflow component. Both will benefit from the current trends in AI. This isn't discussed in detail but it is not hard to imagine potential use cases which I imagine a host of startups are identifying. I see current AI systems as platforms on which to build new applications. Much like AWS is used to build applications today. This will remain true as long as training costs are multiple millions of dollars. I do expect to see unbundling though as we figure out how to get more with less compute.