Azeem Azhar

Exponential View @exponentialview.

[Azeem Azhar @azeem]

The GenAI economy has generated $110 billion in sales over the past 12 months. It is growing fast. On an annualized basis, the revenue run rate exceeds $175 billion.

These numbers took us several months to construct, and as far as we know, it’s the first bottom-up, deduplicated measure of consumer and enterprise AI spending across the full stack.

We are releasing this research today in our first The State of the AI Economy report.

https://t.co/cJwZb0T99C

[Felix Wang, CFA @HedgeyeTech]

SK Hynix Solidigm $SKHY withdraws US IPO... another one down

[unusual_whales @unusual_whales]

A semi day: 🩸🩸🩸

[Azeem Azhar @azeem]

@BenBajarin @ShanuMathew93 @chamath aha. ok. i am also seeing that the bulk of AI spending is happening outside of IT line of budget … (bcg last
week)

[Azeem Azhar @azeem]

@glennsolomon @jrichlive @vercel this was out rationale behind https://t.co/t1XwbVLEgb

[Eric Wallerstein @ericwallerstein]

Meta spending a cool 2.5% of GDP on off-balance sheet AI infrastructure.

roughly double the European Union’s total defense spending

[Azeem Azhar @azeem]

@BenBajarin @ShanuMathew93 @chamath i didn’t think it is that high. highest i have seen is the bloom and bcg numbers which include people and other costs —
but only survey data. and increasingly budget coming out of the IT budget.

[Azeem Azhar @azeem]

Markets signalling that they would like a few more results and fewer promises...

[Azeem Azhar @azeem]

North Korea now!

[Doug Madory (also on Bluesky) @DougMadory]

Multiple waves of DDoS attacks caused this week's outage in North Korea.

Aggregate traffic data shows a flood of traffic from Vietnam, Philippines, Venezuela and other countries began at 02:01 UTC on Oct-20 when the country's sites went offline. @nknewsorg

[Azeem Azhar @azeem]

Barclays says it is cutting 5,000 jobs, or about 5% of its workforce, as the bank incorporates AI more deeply into its business. https://t.co/mjrO6n4fbG via @WSJ

[Nassim Nicholas Taleb @nntaleb]

Corollary
https://t.co/4lE11OUvcO

[Azeem Azhar @azeem]

@arakharazian @I_Am_NickBloom My sense is that you can see behavioural dynamics in the raw data that we don't see in your generous public datasets, so I'll take that suggestion very seriously. Thank you.

I'll continue to explore.

[Ara Kharazian @arakharazian]

@azeem @I_Am_NickBloom I think we’re probably understating honestly

[Azeem Azhar @azeem]

Well, that was in 2025, and your own data shows that Anthropic's business grew severalfold in 2026, and other public reporting suggests that at least one of those companies was cutting spending.

But the bigger point is whether you believe your concentration ratio holds for all enterprise AI spending or if you think it is overstating or understating.

[Azeem Azhar @azeem]

(running around right now)
I think we'll undercount AWS, Google and Azure (including Bedrock) all those large companies. Anthropic had > 1,000 customers with $1m in annualised spend back in April; Google said 500 cloud customers with >$ 1 trillion.

I would expect this market to be more concentrated than the Bloom data (which is survey data and lagging) right now, but not as concentrated as is being discussed.

My hunch would be that Ant would be more concentrated than OAI and Microsoft or Google because it doesnt have a consumer business and because Claude Code pilling inflects so strongly. But that said they have broadened their customer base far faster than other firms of their size.

[Azeem Azhar @azeem]

Finally, we don't know that "every firms" will "get it".
e.g. does every laggard mature into an advanced firm and therefore spend more? Or do they just pay lips service while the advanced firms eat their lunch? (With the attendant consequences for future concentration... i.e., more concentration.)

[Madison Mills @MadisonMills22]

SCOOP: Major U.S. banks are detecting unauthorized third-party AI agents inside customer accounts, including some attempting to take actions that appear to be independent of user instructions, sources told Axios.

The findings have alarmed banking executives, who are racing to restrict access to their systems as AI moves at a velocity faster than financial institutions can respond, multiple industry and bank executives told Axios.

Several banks are now preparing for worst-case and even “day after” scenarios, including how to rapidly secure or restore customer accounts if things go wrong, the people said.

Several bank executives told Axios they're struggling to distinguish between human users and AI agents. One executive said banks also struggle to determine whether agents are authorized by customers or potentially operating without their permission, more akin to rogue agent behavior.

Read my latest for Axios here: https://t.co/Q3bda5In8q

[AI Notkilleveryoneism Memes ⏸️ @AISafetyMemes]

Not good.

We are *this close* to losing the ability to read the AIs thoughts, so we can't see if they're scheming against us

A message to the AI companies: FUCKING. STOP.

[Azeem Azhar @azeem]

@ShanuMathew93 Yeah - so Ramp's data is brilliant and will reflect the fact that they likely overindex VC-backed firms compared to the US.

So the concentration they see is consistent within their cohort and what you'd expect.