Goldman Sachs, Morgan Stanley and the Bank for International Settlements (BIS) are pointing to the same structural shift: artificial intelligence is no longer competing on computing power alone.
The next phase of AI will be shaped by capital. More precisely, by who can secure long-term financing at the lowest cost.
Part I showed how AI has entered the AI Debt Era. Part II examines what follows when financial markets begin to determine the speed, scale and durability of AI infrastructure investment.
Capital Becomes the New AI Advantage
Goldman Sachs projects that global AI and data center investment could reach approximately $5.3 trillion between 2025 and 2030. Morgan Stanley estimates that capital expenditure by Alphabet, Amazon, Microsoft and Meta alone could approach $700 billion this year.
These figures show more than rising investment. They reveal how AI capital is concentrating around a small group of hyperscale companies with the strongest balance sheets and the lowest borrowing costs.
The mechanism is powerful. Cheaper funding allows companies to build larger infrastructure faster. Larger infrastructure strengthens market leadership. Market leadership attracts more capital.
This cycle is widening the gap between AI leaders and the rest of the industry.
AI competition is still about engineering. But it is now increasingly reinforced by financial strength.
BIS Sees AI as a Financial Stability Issue
The growing dependence on debt financing explains why the BIS has begun viewing AI through the lens of financial stability.
In its latest Annual Economic Report, the BIS acknowledged AI’s potential to raise long-term productivity. At the same time, it warned that the scale and speed of AI infrastructure investment could create new vulnerabilities in global financial markets.
The concern is not AI itself. It is the combination of leverage, long investment cycles and uncertain future cash flows.
If financing conditions tighten, interest rates remain elevated or electricity costs rise faster than expected, returns on AI infrastructure could fall short of investor expectations.
A technology investment cycle can become a financial risk when too much capital is deployed before revenue models fully mature.
That is why central banks are now watching AI as a macro-financial issue, not simply as a technology story.
Korea Enters the Global AI Capital Race
Against this backdrop, President Lee Jae-myung’s newly announced AI investment strategy, centered on Samsung Electronics and SK hynix, carries significance beyond domestic industrial policy.
The initiative arrives as AI leadership is being redefined by the ability to mobilize long-term investment across semiconductors, electricity, advanced manufacturing and digital infrastructure.
Korea’s challenge is no longer limited to producing faster chips. It must also secure reliable power, strengthen transmission networks, expand AI data centers and attract the capital needed to finance them.
AI policy is becoming industrial policy, energy policy and capital strategy at the same time.
For Korea, this is a decisive moment. Samsung Electronics and SK hynix remain central to the global AI semiconductor supply chain. But the next stage of competition will also depend on whether Korea can build the infrastructure and financing ecosystem around them.
AI Expands Beyond Technology
Every hyperscale AI data center requires far more than advanced semiconductors.
It depends on electricity generation, power grids, cooling systems, optical communications, land, construction capacity and institutional financing.
As a result, the economic impact of AI is spreading beyond chipmakers and cloud companies.
Utilities, nuclear energy developers, grid operators, cooling technology firms, telecommunications providers, infrastructure funds and pension investors are becoming part of the AI economy.
The AI value chain is no longer confined to technology. It is becoming a full industrial system supported by capital markets.
Capital Becomes AI’s Ultimate Infrastructure
The first AI race was about algorithms.
The second was about semiconductors.
The third is being fought over infrastructure.
The fourth may be decided by the capital itself.
Financial markets are no longer merely funding artificial intelligence. They are shaping how fast it can grow, which companies can compete and which countries can lead the next industrial cycle.
The next generation of AI leaders may not be defined by computing power alone. They may be defined by the ability to finance long-term innovation at the lowest cost.
Today’s Signal
The next AI superpower may not be the country that builds the fastest chips. It may be the one that mobilizes long-term capital more efficiently than anyone else.
Series Conclusion
Part I showed that AI has entered the AI Debt Era. (https://www.koreaittimes.com/news/articleView.html?idxno=154807)
Part II shows why capital markets, central banks and national investment strategies are becoming as important to AI leadership as semiconductors and algorithms.
The future of artificial intelligence will not be written by technology alone.
It will also be financed.
Korean version available here.
