So… the AI Complex has created a zero sum game between themselves and the immediate economic well being of not-immune-to-inflation-and-interest-rates people?
AI Complex vs Main Street
Chamath: The AI Boom Has a Borrowing Problem
“The entire United States economy is levered to the AI complex.
So very narrowly, we're talking about seven or eight companies that set the tone for the entire country. And whether we like it or not, that is where we are.
Why do I say that? The most important companies in the world are in a massive spending binge, right?
These were highly profitable, highly cash accretive, highly cash flow generative businesses that have said, ‘We're going to put the foot down on spending because we believe in what's coming into the future.’ That's what drives GDP and it's what drives growth.
Now, as they do that though, they deplete their cash and they have to raise debt.
The AI complex now owes $340-400 billion to the market. And you'd say, ‘Okay, why is that important?’
The problem is this: These companies are raising so much money it starts to crowd out the capital markets.
And the most important part of the capital market that it starts to crowd out, is the appetite for government treasuries.
And what you can see here is that when you look at these AI firms, the amount of money that they need to raise through the end of '27, it's 30% of all of what the Treasury's about to issue in that same period.
So, whether we like it or not, they are going to start competing for demand.”
CI is going to turn into a feature that everyone has to have?
Which implies that CI as a category breaks down because there is less standalone CI in the world and more CI as a feature.
Introducing Vercel CI
• Intelligently skips tasks with Turborepo
• 2-6× faster CI in our internal repositories
• Optimized for agents via Vercel MCP and CLI
After nearly 8 million tasks, we're excited for the future of agentic CI. In beta on all plans.
https://t.co/IcsuvweAhB
Still relevant prompting in 2026:
if a task is out of your expertise area, first go to ChatGPT and ask,
"Who are three real life people that would have good input on: <x>"
Then feed it into your agent "Do this as <person>"
They need something in latent space to latch onto.
@aronchick Unlimited
Time
We’re not buying “intelligence”
We’re buying time
How much are you willing to pay to bring any specific future into the present?
The best way I've found to think about what is possible with AI is: what if you could brute force everything and were not limited in time or effort? It was just a matter of money.
And then someone did. Everything.
So much fretting from scientists and researchers is just about not wanting to acknowledge that they were a cog in a brute-forcing machine moving very slowly through time.
Why would society choose to be limited by their (or mine or yours) time, if we didn't have to be?
The best way I've found to think about what is possible with AI is: what if you could brute force everything and were not limited in time or effort? It was just a matter of money.
And then someone did. Everything.
@kimmaicutler Being willing to spend on it
Quest can afford them
Guaranteed they don’t spend the enough
And they always choose some other priority over security at every juncture
Anyone who’s ever connected a SCADA system to a network that can be reached, no matter by how many hops through how many layers of security things, will experience deep regret in the very near future
Market has always punished spend on cost centers, rewarded shipping bad software + not fixing.
Bad software has much better margins than good software.
If you’re beholden to that structure, your product + customers are sitting ducks.
Until there’s an ideological break.
7/ The fix side has to run at swarmy machine speed too. Engineer attention is the scarce input, so the design question is where a human actually changes the outcome. Agents reproduce, patch, test and ship. Engineers should see the smol handful of cases that need a judgment call.