Federal Reserve Chairman Kevin Warsh left an entire statement to artificial intelligence at Jackson Hall's annual meeting. Instead of giving interest rate signals in this part, he focused on AI ' s impact on economic growth, business investment and monetary policy judgement.

Capital expenditure growth points to AI construction

Warsh said that the “long-term stagnation” judgement, which had prevailed for too long since the financial crisis, was failing. According to him, corporate capital expenditure grew by about 9 per cent per annum in almost four quarters, the fastest since 2021, more than half of which was related to AI infrastructure. He was more concerned about whether growth was continuing to accelerate than the scale of expenditure itself.

The annualization of head labs is over $100 billion.

Warsh says that AI is moving faster than its most optimistic supporters expected a few years ago. In his view, higher potential growth was emerging.

He also gave a set of market data: the annual token sales of two head AI laboratories exceeded $100 billion, an increase of more than 500 per cent over a year ago. In his presentation, users have become an important source of income for the commercialization of AI by purchasing token models.

  • Corporate capital spending grew by about 9 per cent in the last four quarters.
  • More than half of the increase relates to AI construction
  • Two head labs are annualized token sales over $100 billion.

The Fed considers AI as a new variable

Warsh said that the Fed was following closely the evolution of AI's ecology and had considered AI a “new variable” and possibly even a new production factor. This statement means that AI is no longer just a subject of the technology industry, but may affect the affordable growth of the economy and the level of interest rates.

He raised a number of outstanding issues, including whether AI would continue to push up the productivity of society as a whole, when such changes would occur, and whether token would use supplementary or alternative labour. Warsh did not provide an answer, but merely indicated that the Fed's internal working group on “productivity and employment” was working on these issues.

Add value or continue concentration at a few points

Warsh also mentioned that the benefits of AI would not necessarily be evenly distributed in the chain. A key issue is which scarce asset holders, such as AI Laboratories, Chip Manufacturers, Energy Enterprises and Cloud Service Providers, will have more of the added value.

The article also cites the latest British financial report as a practical reference. Weeda announced a quarterly record of $96.2 billion and revealed that future AI infrastructure-related commitments would be $366.0 billion. This means that AI investments are still being concentrated rapidly on a few core points.

Additional information:The article also cites an earlier report stating that about 95 per cent of the generating AI companies are failing, in contrast to the trend of head chips and cloud infrastructure enterprises to absorb more revenues.