I feel like the implied question isn't of much value on its own. It's all about "where are you going with this?"
Leaving aside the connective tissue between creating value, creating market cap, how these translate to reward/money... What's left?
The subsequent discussion about dividing loot is one discussion. The discussion about making more unicorns might be totally unrelated. The latter is theoretically empirical, but in practice pretty speculative.
My take on these is usually about bottlenecks. A screw holding together a $100bn f35 is still usually cheap. Its importance doesn't determine price, because without scarcity it cannot be a limiting factor. A patent protected screw could, in theory, be priced to reflects value to the buyer/aircraft. A custom designed screw might also be expensive... perhaps reflecting the time value or negotiating power of it's custom maker.
In all cases the plane does not work without the screw.
If we're reviewing costs and such, the above thoughts on "value" might have value. They may even hold clues to cost efficiency more broadly. They may not be too relevant to a discussion about aviation mechanics.
I (respectfully) think pg's pov is too abstract, as is the proverbial opposing pov that he's responding too.
Look... Investors tend to think the bottleneck is "great founders," the modern version of investment opportunity. A lot of founders tend to think that it's investors. Perhaps the bottleneck is corporate and government procurement. Military people and other tribes tend to think the bottleneck is leaders. Economists think bottlenecks are institutional. All convenient viewpoints for the people who hold them.
I don't really understand how automation relates. Unless we're talking about founder-only companies... Why is this points in favour of founder contribution, relative to the (now fewer) employees? Maybe it means that capital's value has risen. Automation is "capital" technically. You might even claim that the public is responsible for more of the value, considering the importance of public domains to technology.