These analysts say the AI spending boom is "not too big." Here’s why.

Recent weeks have seen a substantial increase in AI industry investments, exemplified by OpenAI's $300 billion deal with Oracle and $100 billion from Nvidia, alongside other hyperscalers boosting AI capital expenditures. While these announcements fuel optimism, concerns are rising about the sustainability and circular nature of these transactions, drawing comparisons to the dot-com bubble. However, Goldman Sachs analysts argue that the current AI capex is sustainable, driven by productivity enhancements from AI applications and the growing demand for computational power, noting that current AI spending, at under 1% of U.S. GDP, is significantly lower than prior technology cycles.

Analysis

The artificial intelligence sector is experiencing a significant surge in capital expenditure and strategic partnerships, exemplified by OpenAI's $300 billion deal with Oracle and a $100 billion investment from Nvidia, alongside other hyperscalers. This intense activity, while fueling optimism for continued AI growth, has simultaneously raised concerns among observers regarding the sustainability and circular nature of some transactions, drawing comparisons to the late 1990s dot-com bubble. This mixed sentiment is reflected in the overall moderately positive sentiment score of 0.5 despite underlying optimistic tone.

Goldman Sachs analysts, including Joseph Briggs, counter these concerns, asserting that the anticipated levels of AI investment are sustainable. They cite two primary drivers: AI applications are demonstrably boosting productivity, and unlocking these benefits requires substantial computational power due to rapidly growing model sizes, which outpace reductions in computation and energy costs.

Furthermore, Goldman Sachs highlights that current AI spending, at less than 1% of U.S. GDP, is significantly lower than the 2% to 5% observed during prior large technology cycles. The sustainability of this capex is contingent on a solid broader economy, companies prioritizing first-mover advantages, and investments in compute capacity translating into improved model performance and the emergence of artificial generative intelligence, though the eventual winners remain unclear.

Key Decisions for Investors

• Investors should closely monitor the tangible productivity gains from AI deployments and the actual return on capital expenditure by hyperscalers, rather than solely focusing on headline investment figures.
• Evaluate the underlying economics of AI partnerships, particularly those involving semiconductor firms effectively backing their customers, to discern genuine market demand from potentially circular funding structures.
• Consider a balanced approach to AI exposure, focusing on companies demonstrating clear competitive advantages and sustainable business models within the AI ecosystem, while being mindful of potential overvaluation risks in certain segments.