While tech valuations soar, a hedge fund founder has warned that the $3 trillion AI datacenter boom is being built on “very quickly depreciating assets.” Harris Kupperman’s analysis suggests these datacenters will depreciate twice as fast as the revenue they generate, creating a financial time bomb for the lenders funding them.
This warning is particularly relevant to the $1.5tn “funding gap” being filled by “private credit.” Analysts like Gil Luria of DA Davidson note that these “shadow banking” lenders, “eager to deploy capital into AI,” are “improperly assessing the risks” of this “unproven category.”
These lenders are funding “speculative assets” that may have no customers. This is happening while an MIT study shows 95% of businesses are getting zero return on AI pilots, questioning where the “lofty revenue” will come from to cover the rapid depreciation.
If the assets are depreciating twice as fast as revenue, the $1.5tn in debt taken on to build them could be left underwater, leading to mass defaults. Luria warns this “influx of debt capital… could end up representing structural risk to the overall global economy.”
While “hyperscalers” like Google and Microsoft have the cash flow to absorb this depreciation, the “speculative” half of the market does not. The $3tn boom could be brought down not by a lack of hype, but by a simple, brutal accounting of asset depreciation.
