every job will turn into explaining your intentions to ai
explaining what you want to ai is surpringly time consuming, coders already spend 80% of their time doing it, and this will be true for everyone
I'm finally reading Dune. This quote, which is in the first few pages, hits hard:
"Once men turned their thinking over to machines in the hope that this would set them free. But that only permitted other men with machines to enslave them."
Second for second, @tylercowen packs more substance into a talk than anyone I'm aware of. This is a clear, non-hysterical, and somewhat soothing discussion of our AI future.
NVIDIA IS BUYING ITS OWN CHIPS AND CALLING IT REVENUE
And your retirement account is secretly holding the bag.
This scheme is literally straight out of the Enron playbook...
In January 2026, a special purpose vehicle called Valor Compute Infrastructure was created with one purpose:
Buy Nvidia's chips so Nvidia could book the sale as revenue.
Valor raised $5.4 billion and purchased over 100,000 of Nvidia's GB200 GPUs.
But $1.9 billion of that money came FROM Nvidia itself.
Nvidia invested $1.9 billion into the shell company, then sold that same shell company $5.4 billion worth of its own chips and booked every dollar as revenue.
It's the Girl Scout whose dad bought all the cookies and then she wins the sales contest because Dad was the customer. Except this Girl Scout is a trillion-dollar company and the cookie sale is $5.4 billion.
But it gets MUCH worse:
The remaining $3.5 billion in financing came from Apollo Global Management. Apollo structured the debt, packaged it into securities, and then sold those securities to Athene.
And guess who Athene is? Apollo's OWN insurance subsidiary. The one that sells fixed annuities to American retirees as safe, conservative retirement products.
Follow the chain:
Nvidia funds a shell company with $1.9 billion. The shell company buys $5.4 billion in Nvidia chips. Apollo finances the remaining $3.5 billion. Apollo sells the debt to its own insurance arm. That insurance arm packages it into annuity products and sells them to retirees who think they're buying something safe.
The retirees have no idea that their retirement savings are now backed by 100,000 computer chips sitting in some data center that will be worth pennies on the dollar in three years.
Now look at what's happening inside Athene:
$74.2 billion in US reserves but $217 billion in assets have been shifted to a Bermuda-based captive insurer, outside normal US regulatory oversight.
$103 billion of that portfolio (roughly 35%) is classified as Level 3 assets. That means there is no observable market price.
These assets are valued by internal models, not by actual markets.
And sitting on top of all those unpriced assets? 16.6x leverage.
If you're getting flashbacks to 2008, you should be.
Back then it was mortgages bundled into securities that nobody understood, sold to investors who had no idea what they were holding, rated as safe by agencies that never looked under the hood.
Today it's GPU-backed securities. Computer chips bundled into structured credit instruments, routed through an offshore insurance subsidiary, and sold to you as a retirement product.
The collateral is 100,000 GPUs leased to a single customer through an xAI subsidiary. If xAI stops making lease payments for any reason - financial distress, a pivot in strategy, anything - the entire structure unravels.
And Nvidia releases new architectures every year, so each generation delivers dramatically more compute per watt. A 5 year lease on technology that's obsolete in 2 years creates a mismatch that should terrify every annuity holder in America.
Every single step in this chain is technically legal. The SPV is legal, the lease is legal, Nvidia's equity stake is legal, the securitization is legal, and the Bermuda transfer is legal.
But legality and legitimacy are not the same thing.
I've seen every trick Wall Street has ever pulled in my 45 years of doing this.
And what I'm looking at right now is a pipeline that takes AI infrastructure risk, launders it through 8 layers of financial engineering, and deposits it in the retirement accounts of Americans who never agreed to fund Elon Musk's data centers.
In 2008 it was mortgage-backed securities.
In 2026 it's GPU-backed securities.
Different asset. Same greed. With the same ending.
महाजनस्य संसर्ग: कस्य नोन्नतिकारक: ।
पद्मपत्रस्थितम् तोयम् धत्ते मुक्ताफलश्रियम् ॥
For whom is the company of great people not beneficial? Even a water droplet when on a lotus petal, shines like a pearl. #Sanskrit#Subhashita
Documenting the headwinds I now see for AI.
It won't seem like it, but I love AI and am long-term positive. But when "math doesn't math" I take note.
1. The core thesis for foundation model lab investment has been high upfront investment made worthwhile by significant long-term profits.
2. These are capital intensive businesses and the compute commitments are very high relative to revenue and require strong growth over long time periods. The "leverage" (commitments versus revenue) is extremely high.
3. The fundamentals are not as positive as they previously were:
• Input costs are higher (commodities, chips, power)
• Interest rates are higher
• Competition is more intense
• Scaling Laws are now problematic: exponential costs/power cannot continue
4. Forecasting compute spend is challenging and high risk due to (a) revenue uncertainty and (b) algorithm uncertainty
5. Revenue growth appears to be slowing. The technology is valuable, but ROI is proving to be more expensive and take longer than anticipated.
6. The future is likely "different models for different use cases" with the lower end of the market being highly competitive.
7. Core use cases such as agentic software engineering are likely to need approaches beyond next-token prediction. They are Σ₂ᴾ complexity problems requiring multi-objective optimization and likely a combination of Transformers and other methods.
8. Current forecasts in memory makers are built largely on quadratic attention. That will not persist: we are already seeing work from DeepSeek, Minimax and Nvidia that can cut RAM needs by 80% or more.
9. This means semiconductor valuations are substantially overinflated and will go through the traditional glut versus shortage cycle.
10. For foundation model providers: lower costs with competitive differentiation is good. However, lower costs with a lack of differentiation would mean lower revenues. This makes it harder to (a) service commitments and (b) pay back investors.
11. Leverage is substantially higher than in previous cycles, evidenced by leveraged ETFs, call option activity and margin loans. Korea is particularly susceptible.
12. 0DTE options create a profile that has stronger parallels to portfolio insurance and 1987 than any other point I can remember.
13. The combination of exponential increases in call activity coupled with the ties of semiconductors to structured products means there is a non-trivial systemic risk to the financial system.
14. Implied earnings growth rates are inconsistent with other periods in history.
15. Macroeconomically we cannot and should not fund exponential cost increases. History has shown us repeatedly that there are better ways (see Quick Sort and Simplex).
16. Significant supply is hitting the market via IPOs.
––
Taken together: costs and competition are increasing while revenue growth is likely slowing. Valuations are fragile and prone to technology disruptions that are already here. Systemic financial market risk is extremely high.
1/6
Financial Times: "A company-level OECD analysis of government subsidies across 15 key industrial sectors found that nearly 60 per cent of Chinese firms’ global market share gains since 2005 could be attributed to subsidies."
https://t.co/o2mcZeaagi
🚨Michael Burry just said Elon Musk and Nvidia's deal is built on fake numbers.
Burry published a detailed breakdown calling the entire structure "Fugazi", his word for fake.
He is alleging that billions of dollars in Nvidia chips are being hidden off balance sheets, and that American retirees are unknowingly funding the whole thing.
Nvidia, the world's largest AI chip company sold $5.4 billion worth of its most advanced GPUs, the GB200, to a company called Valor.
Valor is not a real operating business. It is a special purpose vehicle, a shell company created specifically to hold these chips and nothing else. Nvidia also invested $1.9 billion of its own money directly into Valor on top of the sale.
Those 100,000+ chips are now physically inside xAI's data center. xAI is Elon Musk's artificial intelligence company, the one that builds Grok. xAI is using every single one of those chips right now to run its AI models.
But here is what Burry is flagging.
Neither Nvidia nor xAI owns those chips on paper. Valor, the shell company holds legal title. That means $5.4 billion in GPU assets do not show up on Nvidia's balance sheet as inventory.
They do not show up on xAI's balance sheet as assets. They are legally invisible to both companies.
Nvidia gets to book the $5.4 billion as a completed sale and record it as revenue. xAI gets full use of the chips without owning them. And the risk disappears into a shell company in the middle.
Now here is where American retirees enter the picture.
Valor needed $3.5 billion in debt to fund this structure. Apollo provided it. Apollo is one of the largest asset managers on earth with $1.03 trillion under management and $834 billion specifically in private credit.
Apollo raised the $3.5 billion, packaged it into debt securities, and sold those securities to Athene.
Athene is Apollo's own insurance company. It sells fixed and indexed annuities, retirement savings products, to ordinary Americans.
When a retiree buys an Athene annuity, they believe their money is sitting in safe, stable investments. That money is now inside a structure funding Elon Musk's AI data center.
The numbers inside Athene are most alarming.
Athene holds $74.2 billion in reserves. It has moved $217 billion in assets into a captive insurer based in Bermuda, meaning those assets sit outside normal US insurance regulation and oversight.
Of the entire portfolio, 34.7%, equal to $103 billion, is classified as Level 3 assets.
Level 3 is an accounting classification that means there is no observable market price for these assets. No outside party can independently verify what they are actually worth.
The leverage sitting on top of those unpriced assets is 16 times.
Burry's says:
Every step of this structure is technically legal and publicly disclosed. But the entire thing was deliberately engineered across 8 to 12 steps to move credit risk off balance sheets and away from any market pricing.
- Nvidia books the revenue.
- Apollo collects the fees.
- xAI gets the computing power.
- And retirees sitting at the bottom of a 16x leveraged Bermuda insurance structure, holding $103 billion in assets with no market price carry the risk without knowing it exists.
Was in Baroda yesterday
Earlier in Ahmedabad
Prior to that Chennai
The manufacturing revolution
The green energy revolution
The logistics revolution
This is all for real
World class manufacturing facilities by hitherto unknown founders
US / global accreditions
Global manufacturing sourcing from India
Does not matter what the external environment is
Growing strongly and methodically
Remarkable focus on automation
It’s v v different from looking at the exchange screens the whole day and deriving conclusions on the economy from that
The INR depreciation is the best thing that’s happened
Exports become more competitive
Nominal growth accelerates
Outsourcing from India gets turbo charged
As scale rises , backward integration accelerates
Focus on new technologies goes up . We saw a hydrogen electrolyser powered green galvanising plant yesterday .. all being done at “Indian costs”
The new breed of entrepreneurs are actually not new .. they have been at it for 20-30 years and finally now getting the platform to grow aggressively
👆from a @PineTreeMacro investor.
When asked about India's contributions to science, many go back thousands of years and speak about zero, ancient universities, metallurgy, or Ayurveda.
While those achievements are undoubtedly important, India's contributions to modern science are equally remarkable and far less appreciated.
What is fascinating is that many of these contributions emerged during the late nineteenth and early twentieth centuries, when India was under colonial rule and had very limited scientific infrastructure.
Consider a few examples.
Jagadish Chandra Bose demonstrated millimetre wave wireless communication and developed one of the earliest semiconductor detectors. More than a century later, millimetre waves are at the heart of modern 5G and future 6G communication systems, while semiconductor devices underpin the entire digital economy.
Meghnad Saha developed the ionization equation, which made it possible to determine the temperature and chemical composition of stars from their spectra. Modern astrophysics, stellar evolution studies, plasma science, and even fusion research continue to rely on principles that emerged from his work.
Satyendra Nath Bose developed the statistical framework governing an entire class of particles now known as bosons. Lasers, superconductors, Bose-Einstein condensates, photonic technologies, and many emerging quantum computing platforms trace their origins to ideas he developed in a few pages written in Dhaka.
Upendranath Brahmachari developed Urea Stibamine, one of the first successful targeted treatments for a major infectious disease. Beyond saving countless lives from Kala Azar, his work helped establish the foundations of modern antiparasitic chemotherapy and drug discovery for neglected tropical diseases.
Prafulla Chandra Ray's contribution went far beyond discovering new chemical compounds. He demonstrated that scientific research could be translated into indigenous industrial capability. Through Bengal Chemicals, he laid some of the earliest foundations for science-driven manufacturing in India, a legacy that can be seen today in India's globally significant pharmaceutical industry.
These were not incremental advances. These were foundational contributions that shaped entire scientific disciplines and industries worth trillions of dollars today.
These foundational works also helped many receive Nobel Prizes later.
India has never lacked talent. Even under colonial rule, Indian scientists produced ideas that changed the world.
The challenge now is to build institutions that trust scientists, reduce bureaucratic friction, invest consistently in research, protect academic freedom, and create environments where transformative ideas can flourish.
If a colonized nation with limited resources could produce all this great science, one can only imagine what India can achieve if we fully unleash the talent that exists today.
We just need to empower 'real' scientists.
https://t.co/DsttYZc9Ld