Nicolai Tangen, CEO of Norges Bank Investment Management pressed IBM CEO Arvind Krishna directly on whether AI is a bubble (Save this).
And Krishna responded with what has become known inside financial circles as the $8 trillion math problem.
A single gigawatt of AI data center capacity filled with accelerators, liquid cooling, and power infrastructure costs roughly $60 to $80 billion to build and populate.
The industry has committed to more than 100 gigawatts of buildout globally.
That is $6 to $8 trillion in capital expenditure and because AI grade hardware depreciates on a five-year cycle, that entire sum must be effectively replaced and refreshed every five years.
To service the interest on $8 trillion in capital at a conservative 10% borrowing rate, the AI ecosystem would need to generate approximately $800 billion in annual profit, a number that currently exceeds the combined net income of every large technology company in the world.
Goldman Sachs estimates $7.6 trillion in aggregate AI CapEx between 2026 and 2031 alone, and Reuters Breakingviews has flagged that even if the capital is available, physical bottlenecks power permits, land, cooling infrastructure, and electrical grid connections mean that half of the planned data center projects are being cancelled or delayed before they ever go live.
Krishna also raised a second, structurally distinct concern that markets have largely ignored.
He argued that the largest foundation models, GPT, Gemini, Claude, Llama are converging toward commodity status.
When a product is a commodity, switching costs collapse.
When switching costs collapse, pricing power evaporates and margins compress regardless of how much capital was spent building the capability.
Morningstar's equity research team conducted a review of 132 technology companies in 2026 and found that AI had caused moat rating downgrades across roughly 40 major stocks concentrated in enterprise software, IT services, and SaaS with Adobe, Salesforce, Workday, and ADP among the companies whose competitive moats have materially weakened.
The implication is that the companies spending the most on AI model development may be building an asset that is simultaneously the most expensive to produce and the most difficult to monetize with durable margins.
This bear case is serious but it is also incomplete and that is what makes Krishna's framing so important to understand precisely.
When pressed further, Krishna explicitly said he does not believe there is an AI bubble in the technology itself only in a subset of the infrastructure capital that is being deployed against speculative assumptions rather than proven demand.
He draws the same analogy, the fiber optic overbuild of the late 1990s. Dozens of companies went bankrupt laying cable that nobody was using.
And yet that exact "wasted" infrastructure became the physical backbone of every cloud company, every streaming service, every mobile network, and every modern AI training cluster that followed.
The builders lost, the infrastructure won.
And the companies that were built on top of it, Amazon, Google, Netflix, Salesforce compounded for two decades.
The question, as Krishna framed it, is not whether AI is real.
It is which capital deployment earns a return versus which gets stranded and crucially, whether you own the stranded assets or the companies built on top of them.
On winners, Krishna was direct that distribution is the moat on the consumer side, and enterprise is wide open.
The data supports this, Meta with 3.3 billion daily active users across Facebook, Instagram, and WhatsApp is building AI into a distribution network that no startup can replicate at any cost.
Meanwhile, the productivity evidence arriving in real time is beginning to challenge the bear case's revenue projections.
Jensen Huang just showed on stage at Computex that GitHub commits, the universal measure of global software output nearly tripled in the first months of 2026, effectively converting $3 trillion in developer salaries into $9 trillion in productive output.
That is measurable, real time economic value already flowing through the system and it feeds directly back into token demand in a compounding loop that Krishna's static CapEx math does not fully capture.
๐จ SOMETHING DOESN'T ADD UP
The S&P 500 just pushed another leg higher.
Everyone is celebrating new highs.
But almost nobody is talking about risk.
Every previous cycle inside this channel ended with a 20%+ dump.
Most people still see strength.
I see a setup that has already played out multiple times.
The crowd always gets bullish at the wrong time.
I publicly called the 2022 market bottom and the 2025 market top before most people saw them.
Follow and turn notifications on now.
Don't be someone else's exit liquidity.
An old trader sits quietly at his desk every morning.
His rules read:
1 good trade a day
3 mistakes and stop
A young trader joins the market.
โThis is easy,โ he says.
The young trader takes one trade. Then a second. Then a third.
After chasing every move, forcing entries, and flipping directions, he stares at his screen late into the day.
Then he turns back, grinning proudly. โOld man,โ he says, โyou only made one trade while I took seven.
Maybe youโre too slow for this market.โ
The old trader smiles and shakes his head.
โFunnyโฆ every time somebody loses focus, they confuse doing more with making money.โ
He points at the screen.
โTrading was never about finding more trades. It was about making fewer mistakes.โ
โEvery leak in your system costs you clarity. Every forced trade costs you alignment. Every emotional decision pulls you further away from the edge.โ
The young trader looks back at his account. The old trader closes his laptop.
โMost people think trading is about action.โ
โItโs actually about discipline.โ $SPY
๐จ The AI ROI numbers are starting to look very ugly.
Even under "best case" assumptions โ assuming zero costs, just revenue against capex โ the Financial Times calculated the implied return on hyperscaler AI investment from 2025 to 2030.
Only one of them clears positive.
Implied return on AI investment (FT / Panmure Liberum)
โ Microsoft: -9.2%
โ Alphabet: -15.7%
โ Amazon: +7.2%
โ Meta: -28.8%
โ Oracle: -35.6%
And remember: that's assuming zero costs. In reality, GPUs depreciate, power bills run, salaries get paid.
The real returns are worse.
This is exactly why the dot-com comparison keeps coming up. Incredible technology does not automatically mean sustainable economics. The internet survived. Most internet companies didn't.
Two anecdotes from this week alone
Vivek Garipalli, Fortune 20 insider: a CEO asked for $1B in AI-driven opex savings this year. The team spent $200M on tokens chasing it. The results? Modest customer service savings and slightly less hiring in engineering. The CEO has now ordered token costs to be dramatically slashed because the ROI isn't there.
Axios: an AI consultant reported a single client spent half a billion dollars in one month after forgetting to put usage limits on Claude licenses for employees.
Right now hyperscalers are spending trillions hoping future demand catches up to present capex.
That's not certainty. That's a leveraged bet.
The technology is real. The infrastructure buildout is real. The eventual winners will be real.
But "AI is transformative" and "every hyperscaler will earn its capex back" are two completely different statements.
In 2000, the internet was real too.
Cisco has recovered.
After 26 yearsโฆ