Apes
You have land taxes and or rent, building upkeep, staffing costs, electricity, water, hardware replacement costs.
And new build DCs have blown all these costs through the roof justifying the decision because the price of compute is so high. When the prices come crashing down, the expenses will remain fixed where they are now.
New datacenter builds are so far along the curve of diminishing returns it's absurd. No one is going to want to pay to run a datacenter that costs 10x as much to run for the same compute.
And the problem is with all these "freed" resources, the entire pipeline will be affected. No one will want to buy any new silicon if they can buy a B200 for $1000. We could potentially see a decade or more of stagnation in the chip sector, or even significant regressions in capabilities as foundries are shut down due to lack of demand.
The impact of what is coming scares me to my core. I don't think we're going to bounce back from this any time soon.
If the cutting edge OpenAI token prices are $80 per 1M token, and the open source tokens are $1 per 1M token, that's a huge gap of "this will never be able to make money under any scenario if the bubble bursts" that will catch a lot of these new datacenters. No one will run a datacenter that costs $5 per 1M token to sell at $1 per 1M token even if the debts are cleared.
National debt through the roof, inflation through the roof, PHD and research programs gutted, non-ai startups dead and unfunded for the last 4 years. These are just a few things that have been sacrificed on the altar of this bubble - there's far more I haven't recounted.
We're already in a widespread long term economic collapse, but the delusion just hasn't broken yet.
That's the theory anyway - our Idiot King and his idiots have completely missed the point of the US military existing and are using it as a primary method of engagement, which is causing the economic boon used to fund the military to evaporate.
As an aside, it's not a huge issue, but China's military costs use different accounting than the US, and seem lower by comparison. Apples to apples, China probably spends about half what the US does on military.
Its the doctors doing this, not the MRI.
There's this weird definition switch that always happens with the "overdiagnosis" defense where the information gets blamed for the overdiagnosis. An MRI doesn't provide any diagnosis in any sense of the definition. A doctor does.
Claiming an overdiagnosis defense is essentually implying the medical industry is worse for most than doing nothing.
You're not sure of whether this is a good idea or not, so you ask various physicians, and the consensus is unanimous: the very suggestion is offensive, do you think doctors are unclean?
A clear conclusion has been achieved.
Mainstream medicine is hyper optimized for the most common 80% of cases. At a glance it makes sense: optimize for the common case. Theres some flaws in this logic though - the most common 80% also conveniently overlaps heavily with the easiest 80%. If most of the problems in that 80% solve themselves, then what actual value is provided by a medical system hyper focused on solving non-problems? The real value from the medical system isnt telling people "it's probably just a flu, let's just give it a few days and see" it's providing a diagnosis for a difficult to identify condition.
So if your question is "how do we maximize value and profit in aggregate for providing medical care to large groups of people", mainstream medicine is maybe a good answer.
But if your question is "how do we provide the best care to individual patients" then mainstream medicine has significant problems.
There's no single points of failure: as you get older, everything just starts wearing out and failing.
If you cure heart disease and cancer, then others will just take their place: strokes, respiratory disease, Alzheimer's disease, falls.
And even if you do extend your lifespan, the reality is quality of life at 90+ is a lot worse than in your 20s or 30s.
A bond with a face value of $1000 means the government has $1000 of debt regardless of what is paid for the bond.
The coupon payments represent the "interest" on that debt - the $20 coupon means the government is paying $20 of interest per year.
Paying below face value doesn’t make the difference "interest." It simply means investors are buying the bond at a discount, so the government receives less cash upfront in exchange for repaying the full $1,000 at maturity. Bonds differ from traditional loans in that their market price can fluctuate, but the debt obligation remains fixed at the face value.
In practice, the government's accounting labels the discount as an "interest expense", so it still gets captured as interest in the budget.
The reason Uber can get away with pretending it's just a "connector" is because the entire tech ecosystem has been allowed to normalize that kind of control without accountability.
Look at Apple and Google: they take a 30% cut on every sale and ban any competing payment systems. That's the same pattern - absolute gatekeeping disguised as "market facilitation."
Our regulators have become so complacent that this behavior is now seen as the default way digital markets work. The problem isn’t just Uber's misclassification; it's that the entire platform economy is built on pretending these companies are neutral middlemen when they're really gatekeepers.
* Tariffs and Trade Uncertainty – Elevated tariffs and rapidly shifting trade policies are raising costs for manufacturers and discouraging hiring and investment (Investopedia; Atlanta Fed survey; CBO analyses).
* Automation and AI Displacement – Automation and AI, especially in low-skill occupations, are reducing new job creation and wages for some workers (academic studies in arXiv and PMC).
* Restrictive Immigration Policies – Tightened immigration and visa processes are straining labor supply, particularly in sectors that rely on immigrant workers (Axios, 2025 labor coverage).
* Small Business Strain from Economic Pressure – Tariff-related uncertainty is leading small businesses to slow hiring or lay off employees (Joint Economic Committee report, 2025).
* Offshoring and Outsourcing Trends – Technological advances are enabling offshoring and automation, substituting domestic labor with remote or machine-based alternatives (academic research in World Development, 2024).
* Tariffs and Trade Uncertainty – Elevated tariffs and rapidly shifting trade policies are raising costs for manufacturers and discouraging hiring and investment (Investopedia; Atlanta Fed survey; CBO analyses).
* Automation and AI Displacement – Automation and AI, especially in low-skill occupations, are reducing new job creation and wages for some workers (academic studies in arXiv and PMC).
* Restrictive Immigration Policies – Tightened immigration and visa processes are straining labor supply, particularly in sectors that rely on immigrant workers (Axios, 2025 labor coverage).
* Small Business Strain from Economic Pressure – Tariff-related uncertainty is leading small businesses to slow hiring or lay off employees (Joint Economic Committee report, 2025).
* Offshoring and Outsourcing Trends – Technological advances are enabling offshoring and automation, substituting domestic labor with remote or machine-based alternatives (academic research in World Development, 2024).
The limiting factor is going to be the VRAM on the 5090, but nvidia intentionally makes trying to break the 32GB barrier extremely painful - they want companies to buy their $20,000 GPUs to run inference for larger models.
Architecturally, the DGX Spark has a far better cache setup to feed the GPU, and offers NVLINK support.