The idea would be that the Paul Graham "brand" has a lot at stake among startups. If the bottom falls out of the market in the next 2 years (or whatever you want to define as "short term"), then he'll come out the other side looking that much better, that he got out in plenty of time. If everything continues on its current trend, he still sleeps perfectly well at night and his reputation hasn't taken a hit at all.
Don't misconstrue this as criticism of anyone. I have nothing whatsoever against either Paul Graham or Sam Altman. This is just my personal opinion on the matter.
Graham wrote a "Request for Startups" that expressed an honest and tradeable opinion: "these 6 categories are likely to make me money, I want to fund them." Now compare to Sam Altman's RFS: "here are all the possible categories, don't be mad at me for forgetting yours, plus a few feel-good platitudes that will make the media like me."
YC probably needs a schmoozer/politician more than an intellectual right now. But that doesn't mean PG wants to be that person.
In any case, I wonder how companies with such massive escalation in their valuation will hire and raise subsequent funding. It really makes me wonder, you look at companies like Uber with $xx billion valuations, what's left for someone who joins now as an employee or investor?
That's always been the case. The opportunity for outsized gains in any risky venture is at the beginning where the risk is the greatest. Where there's no risk, there's no opportunity.
Off the top of my head: startups being seen as a legitimate career path alongside BigCo/academia, greater founder control of boards (almost unheard of post-Series A a decade ago), willingness to fund people without "track records", higher relative value of ideas, etc.
That's silly. He believed in startups, that's why he started YCombinator. That he's no longer at the helm of the startup he created doesn't mean he doesn't believe in startups anymore.
It just means that he's done them and helped others do them enough to where he's ready to move on. I'm a Ruby developer right now. If I switched to Rust tomorrow, that doesn't mean that I found Ruby lacking. Just that I want to explore a new kind of programming. They both fit the bill nicely for the types of problems they were intended to solve.
Don't focus on solutions. Focus on problems. The solution will come out of the problem. The startup 'solution' solves a particular kind of social problem. It still solves that problem. PG is finding a different problem, that's all. Perhaps at the junction of the funding problem and the founding problem, if I were to make a guess.