Nvidia exec: AI compute costs exceed human worker costs currently
Original: ‘The cost of compute is far beyond the costs of the employees’: Nvidia exec says right now AI is more expensive than paying human workers
Why This Matters
Reveals AI deployment economics still favor human labor despite massive corporate AI investments
Nvidia VP Bryan Catanzaro says AI compute costs far exceed employee costs for his team. MIT study found AI automation economically viable in only 23% of vision-related roles. Tech companies spent $740 billion on AI capex in 2026, 69% increase from 2025.
Bryan Catanzaro, Nvidia's VP of applied deep learning, told Axios that compute costs for AI far exceed employee costs for his team. An MIT 2024 study supports this, finding AI automation economically viable in only 23% of roles where vision is primary. Despite higher AI costs, Big Tech announced $740 billion in AI capital expenditures in 2026, a 69% increase from 2025. Uber CTO Praveen Neppalli Naga said budget needs were 'blown away' by AI tool costs. AI software fees increased 20-37% over the past year according to Tropic. McKinsey projects AI expenditures could reach $5.2-7.9 trillion by 2030. Despite unclear productivity benefits, tech layoffs reached 92,000 in 2026 across nearly 100 companies.