Tom Lee is challenging Michael Burry’s Enron comparison as investors weigh roughly $3 trillion in off-balance-sheet commitments tied largely to the AI infrastructure boom.
Key Points:
- Lee says the $3 trillion figure does not show the full financial risk behind AI spending.
- Burry has doubled down on his Nvidia short while criticizing the chipmaker’s planned $500 billion financing push.
- The dispute centers on whether AI productivity can justify a buildout already reshaping corporate finance.
Michael Burry Warning
Burry intensified his warning after targeting Nvidia’s planned $500 billion AI financing push, arguing that some structures resemble financial engineering associated with Enron before its 2001 collapse. The investor, known for betting against subprime mortgages before the 2008 financial crisis, has doubled down on his Nvidia short.
A Wall Street Journal analysis found about $1.2 trillion in leases that have not started and another $1.9 trillion in chip purchase commitments. Those obligations generally stay off balance sheets until leases begin or goods and services are delivered.
“Structuring credit is a natural part of the system. Structuring unnatural credits to prolong momentum late in the bull phase is where the worry comes in,” Burry wrote. Enron’s roughly $60 billion bankruptcy was the largest in U.S. history when the energy company collapsed.
Also Read: Bitcoin Flashes 8 Of 12 Bottom Signals, But History Says Wait
Tom Lee Rebuttal
Lee rejected the comparison during a CNBC appearance with Dryden Pence, arguing that gross financial obligations can far exceed the assets or cash immediately at risk. “The revelations from the journal article are actually helpful, but they’re giving people an incomplete picture of how financial systems work,” Lee said.
Lee also argued that companies can reduce future spending before some commitments reach their balance sheets, while today’s largest technology firms have stronger margins than companies caught in the late-1990s telecom bubble. Pence estimated AI infrastructure already represents 2% to 2.5% of U.S. GDP and could exceed defense spending by 2027.
The bullish case depends on adoption. Pence said only 30% of companies report AI productivity gains and 7% consider implementation complete, leaving room for wider use but uncertainty over returns.
Lee also chairs BitMine, which reported 5.82 million Ethereum (ETH) on Aug. 17, equal to 4.8% of supply. Earlier in August, Burry warned of a possible 1987-style market crash, while Lee argued that AI and tokenization could strengthen Ethereum relative to Bitcoin (BTC).
Read Next: OpenAI Launches ChatGPT For Teens With Stronger Safety Rules
