Would hyperscalers cut back their AI capex?

Key Concepts

• Large Language Models (LLMs): AI models designed to understand and generate human-like text. Concerns center around their cost of development and questionable profitability.
• Hyperscalers: Large technology companies (e.g., Amazon, Microsoft) with massive data centers and computing power, heavily investing in AI.
• Private Credit: Lending to companies by non-bank financial institutions, often with less regulatory oversight and potentially higher risk.
• Momentum Investing: A strategy focused on buying stocks that have shown strong recent price increases, with the expectation they will continue to rise.
• AGI (Artificial General Intelligence): A hypothetical level of AI that possesses human-level cognitive abilities.
• Capex (Capital Expenditure): Funds used by a company to acquire, upgrade, and maintain physical assets such as property, plants, buildings, and equipment.

Market Pullback & AI Investment Concerns

The discussion centers around a recent market pullback, specifically triggered by reactions to OpenAI’s new funding round and performance of key tech stocks. Kim Forest, founder of Bokeh Capital Partners, expresses skepticism regarding the financial viability of large language models (LLMs) despite significant investment. She highlights a “circular” funding pattern where companies like Amazon, Nvidia, and SoftBank are repeatedly investing in OpenAI, with Amazon providing web services to OpenAI while simultaneously funding it. Forest questions whether consumers will ultimately pay the substantial costs associated with developing and maintaining these models, stating, “I don’t trust it because it’s not correct enough for me to trust it… I just don’t see the return for investors in this at in any kind of short way given the billions of dollars that have been spent.” She differentiates between LLMs and “smaller problem solving AI,” expressing enthusiasm for the latter.

Private Credit & Financial System Risks

The conversation also addresses the recent downturn in private credit stocks. Forest connects this to Jamie Dimon’s recurring warnings about “cockroaches” in the financial system, noting his consistently cautious outlook. She explains that while private credit companies should theoretically offer higher shareholder value due to their non-bank structure, they lack the regulatory safeguards of banks, potentially leading to excessive risk-taking. This concern is contributing to a broader selloff in the financial sector, as the financial system is interconnected. Forest has “always been cautious about recommending private anything, but specifically private credit.”

Hyperscaler Spending & Investor Sentiment

The discussion then turns to the continued AI capital expenditure (capex) plans of major tech companies, often referred to as “hyperscalers.” Despite market concerns about aggressive spending, Forest believes companies like Microsoft, Meta, Amazon, and Apple have the financial resources to continue investing. She anticipates continued funding from venture capital firms, who are inherently risk-tolerant. However, she reiterates her skepticism about the ultimate return on investment for these projects, asking, “Who’s paying the bill?”

Nvidia’s Performance & Momentum Shift

Nvidia’s stock decline following its earnings report is analyzed. While the earnings themselves were strong, the market reacted negatively, with shares dropping 4.2% on the day of the discussion. Forest attributes this to investors taking profits after a period of significant gains and recognizing a shift in Nvidia’s growth trajectory. She describes Nvidia as a former “momentum darling” that is no longer delivering the “ridiculous numbers” required to sustain momentum investing. She notes, “It’s not a stock for momentum investors anymore because they’re not putting up those ridiculous numbers that they have in growth.” She suggests the sell-off is driven by momentum traders exiting their positions, rather than a fundamental flaw in the company itself, and that it will “take a while for a once momentum stock to find its place in the world.”

Logical Connections

The conversation flows logically from a broad market overview to specific concerns about AI investment, private credit, and individual stock performance. The discussion consistently links these areas, highlighting the interconnectedness of the financial system and the impact of investor sentiment. Forest’s perspective provides a consistent thread throughout, emphasizing the importance of financial prudence and realistic expectations.

Data & Statistics

• Nvidia Stock Drop: Down 4.2% on the day of the discussion following earnings report.
• Magnificent 7: Reference to the group of large tech companies (Microsoft, Meta, Amazon, Apple, and others) driving market performance.
• Billions/Hundreds of Billions: Repeated references to the massive financial investments being made in AI development.

Synthesis/Conclusion

The core takeaway is a cautious outlook on the current market, particularly regarding investments in AI and private credit. Kim Forest’s analysis suggests that investor enthusiasm may be waning as the financial realities of these sectors become clearer. While acknowledging the potential of AI, she stresses the need for realistic expectations about profitability and the risks associated with aggressive spending. The discussion highlights the importance of understanding the underlying fundamentals of investments and recognizing the shift in market dynamics, particularly the end of the “momentum” phase for companies like Nvidia. The overall message is one of skepticism and a call for a more pragmatic approach to investment in the current environment.