Key Takeaways

  • Chamath Palihapitiya predicts nine to 10 massive conglomerates will dominate technology over the next decade.
  • Winning companies will vertically integrate the entire stack, starting at physical land and electrical power up to consumer applications.
  • Traditional boundaries dividing software from services are disappearing, marked by moves like Palo Alto Networks acquiring an open-source model firm.
  • Public markets are penalizing tech firms that spend free cash flow on share buybacks instead of data centers, frontier models, and strategic acquisitions.

Ten Megacorps Will Control the Full Stack

For twenty years, the standard playbook for tech founders was asset-light software. You rented servers from a cloud provider, hired engineers, and sold digital access at eighty percent margins. You avoided buying real estate, running power lines, or managing industrial equipment.

Artificial intelligence reverses that dynamic. Palihapitiya outlined the emerging structure directly: “I think that there are going to be nine to 10 mega corps that get built over the next decade because of AI.” (play from 0:17) Survival in this new environment requires owning the inputs that make computation possible. Palihapitiya explained the scope of that integration: “And I think that each mega corp is going to go from the application layer all the way down to the dirt.” (play from 0:25)

If you do not own the power generation, the acreage, the server clusters, and the proprietary models, you are merely an intermediary paying tolls to the companies that do. As the All-In hosts observed, “So I think what AI is creating essentially is halves and have nots. The halves are the folks that have these critical assets that can be very tightly vertically integrated.” (play from 1:36) Companies without capital scale will struggle to compete with entities that control both the power grid connections beneath the hardware and the consumer products above it.

Markets Punish Buybacks and Reward Hard Assets

Public markets have already started repricing this shift. For years, mature enterprise software vendors relied on financial engineering to flatter their share prices. When growth slowed, management directed free cash flow toward repurchasing shares rather than building new technical capabilities. That trade is failing.

As the All-In hosts noted, “You know what's not going to be valuable if you take your free cash flow and you spend it on buybacks.” (play from 10:15) Investors want cash channeled into physical assets, specialized data centers, and acquisitions that expand technical moats. Palo Alto Networks illustrated this strategy by acquiring an open-source model firm, erasing the boundary between pure security software and base model infrastructure.

Wall Street no longer cares about legacy definitions. The All-In hosts put it bluntly: “Does anybody actually care whether you're a software company or a services company? You shouldn't. The people that care are morons. What you should care about is what is your free cash flow?” (play from 6:00) The goal is generating capital and spending it on physical capacity and proprietary models before competitors can catch up.

What to Do With This

Audit your unit economics this week to see how much margin you surrender to third-party model providers. If your product relies on generic API calls without proprietary distribution or owned infrastructure advantages, shift your engineering roadmap toward workflows that lock in proprietary customer data.