The ongoing clash between Nvidia and notable investor Michael Burry is intensifying. Burry, known for his role in the 'Big Short,' has vocally criticized the current AI investment surge, likening it to the dotcom bubble of the late 1990s, with Nvidia playing a central role. In response, Nvidia quietly sent out a detailed memo to analysts, directly addressing and attempting to disprove Burry's assertions, his name explicitly mentioned at the outset.
In a thorough seven-page document, Nvidia headlined with a section titled ‘Michael Burry on Twitter / X’ as its primary source of claims to counter. Following the memo's release, Burry stood by his perspective in a Substack post, acknowledging Nvidia's attempt to refute his analysis which critiques topics like stock-based compensation and depreciation. 'I am not claiming Nvidia is Enron. It is clearly Cisco,' Burry stated.
Burry has consistently warned that the current AI infrastructure enthusiasm is more akin to the telecom expansion of the late 1990s than to the dot-com collapse. He highlights significant capital expenditures, extensive depreciation schedules, and rising valuations as signs that the market might be misinterpreting a surge in supply as a sign of ongoing demand.
The memo, as first reported by Barron's, addresses Burry's concerns about Nvidia's stock-based compensation and stock buyback initiatives. Nvidia clarified, 'NVIDIA repurchased $91B shares since 2018, not $112.5B; Mr. Burry appears to have incorrectly included RSU taxes.' They further explained that employee equity grants should be viewed separately from stock repurchase program performance and noted that Nvidia's compensation practices align with industry standards.
Nvidia also countered Burry's depreciation life accusations, suggesting that customers accurately depreciate GPUs over a realistic four to six years. Older models like the A100s, introduced in 2020, reportedly still run at high utilization rates, disputing critics who claim their useful life is overstated.
The memo rebuffs Burry's notion of 'circular financing,' pointing out that Nvidia's strategic investments comprise a minimal portion of its revenue, with AI startups mainly securing capital from external sources.
Burry draws parallels between Nvidia and Cisco of the late 1990s — both seen as pivotal suppliers amid significant capital investment cycles. He claims that just as telecom firms invested billions in Cisco equipment decades ago based on overestimated internet traffic growth, today's large-scale AI infrastructure investments are similarly overambitious. Burry argues that, like in the past, supply might outpace actual demand, with Nvidia now playing the role that Cisco once did during the dot-com era.
The analysis suggests the industry's assumptions about boundless AI demand and the longevity of data center power and GPU usage could be overly optimistic. 'And once again there is a Cisco at the center of it all, with the picks and shovels for all and the expansive vision to go with it. Its name is Nvidia,' Burry concluded in his critique.