Podcast Episode
His comments arrived at a pivotal moment for the tech sector. The four major hyperscalers have unveiled combined capital expenditure plans totalling roughly six hundred and fifty billion dollars for twenty twenty-six, representing a sixty percent increase from the previous year. Amazon announced a two hundred billion dollar budget, the largest infrastructure commitment in tech history. Alphabet revealed plans to spend up to one hundred and eighty-five billion, while Meta projected between one hundred and fifteen billion and one hundred and thirty-five billion dollars.
ServiceNow has fallen over forty percent in three months, while Salesforce declined roughly twenty-five percent and Adobe dropped nearly twenty percent. The broader software sector index is down about twenty percent year-to-date.
Despite the turbulence, Huang emphasised that AI companies like OpenAI and Anthropic are profitable but remain "computer constrained," signalling sustained demand for computing infrastructure well into the next decade.
Nvidia CEO Says AI Buildout Has Seven to Eight Years Left as Big Tech Commits Six Hundred and Fifty Billion Dollars
February 6, 2026
Audio archived. Episodes older than 60 days are removed to save server storage. Story details remain below.
Nvidia CEO Jensen Huang has pushed back against AI spending fears, declaring the industry is in a once-in-a-generation infrastructure buildout with seven to eight more years to go. His comments came as the four major tech hyperscalers unveiled combined spending plans of approximately six hundred and fifty billion dollars for twenty twenty-six, even as software stocks suffered a nearly one trillion dollar selloff triggered by fears of AI disruption.
Nvidia's Huang Defends Massive AI Investment Cycle
Nvidia CEO Jensen Huang has delivered a forceful defence of the artificial intelligence spending boom, declaring that the industry is in a "once in a generation infrastructure buildout" with seven to eight more years to run. Speaking on CNBC, Huang described demand for Nvidia's products as "sky high" and pushed back against fears that AI will replace traditional software, calling such concerns "the most illogical thing in the world."His comments arrived at a pivotal moment for the tech sector. The four major hyperscalers have unveiled combined capital expenditure plans totalling roughly six hundred and fifty billion dollars for twenty twenty-six, representing a sixty percent increase from the previous year. Amazon announced a two hundred billion dollar budget, the largest infrastructure commitment in tech history. Alphabet revealed plans to spend up to one hundred and eighty-five billion, while Meta projected between one hundred and fifteen billion and one hundred and thirty-five billion dollars.
Software Stocks Reel From AI Disruption Fears
The massive spending announcements coincided with a brutal selloff in software stocks, triggered when Anthropic released new AI automation tools targeting legal, finance, and marketing sectors. A Goldman Sachs basket of software stocks plunged six percent on Tuesday, its steepest single-day drop since April, wiping out approximately two hundred and eighty-five billion dollars in market value.ServiceNow has fallen over forty percent in three months, while Salesforce declined roughly twenty-five percent and Adobe dropped nearly twenty percent. The broader software sector index is down about twenty percent year-to-date.
Industry Leaders Push Back
Huang addressed the disruption fears directly at Cisco's AI Summit in San Francisco, arguing that AI's most promising advances stem from its ability to leverage existing software rather than replace it. Private equity executives joined the defence, with KKR co-CEO Scott Nuttall calling the selloff "excessive" and Ares Management CEO Michael Arougheti urging investors to distinguish between software sectors positioned to benefit from AI and those facing genuine disruption.Despite the turbulence, Huang emphasised that AI companies like OpenAI and Anthropic are profitable but remain "computer constrained," signalling sustained demand for computing infrastructure well into the next decade.
Published February 6, 2026 at 8:51pm