Will the AI Bull Market Return to Wall Street? Meta Platforms' New Business Could Be the Catalyst

Key Takeaways:

• Meta Compute could become a significant revenue stream for Meta Platforms, similar to how AWS became crucial for Amazon.
• The sell-off in memory stocks may be overblown, as successful AI compute leasing could enhance the economics of investments in GPUs and data centers.
• If Meta demonstrates that AI infrastructure can generate substantial cash flow, current valuations of major hyperscalers may be too conservative.

Introduction

Meta Platforms has invested heavily in artificial intelligence, yet these expenditures have been perceived as costs. The introduction of Meta Compute, which commercializes its AI infrastructure, signals a potential shift towards monetization. This has led to a sell-off in memory stocks, as investors speculate on excess computing capacity. However, this interpretation may be overly simplistic.

The Potential of Meta Compute

Meta Compute could create a high-margin revenue stream by leasing computing capacity and AI models. According to Morgan Stanley, leasing 250 MW of computing capacity could increase Meta's earnings per share (EPS) by approximately USD 3 by 2028, while a scenario involving 1 GW could boost EPS by as much as USD 11.9, representing a significant portion of the company's projected earnings.

Market Reactions and Misinterpretations

The market reacted swiftly to the news of Meta Compute, leading to a sell-off in companies supplying memory for AI data centers. Investors assumed that Meta's plans indicated a surplus in GPU infrastructure, suggesting lower future demand for memory components. However, this overlooks the possibility that Meta is aiming to optimize the utilization of its existing infrastructure rather than signaling a reduction in future investments.

Long-Term Implications for the AI Industry

Successful execution of Meta Compute could paradoxically support long-term demand for GPUs and advanced memory, as profitable leasing would incentivize further investment in AI infrastructure. The key question remains whether Meta can attract enough customers to monetize its resources effectively.

Competitive Landscape

Companies that rely solely on renting GPU capacity may face significant challenges if major players like Meta, Microsoft, Google, and Amazon enter the market with similar services. The competitive edge of specialized providers could diminish, leading to consolidation within the GPU cloud market.

Investment Outlook

Despite substantial investments in AI infrastructure, leading technology companies are trading at attractive valuations compared to the broader market. The current focus on costs may overshadow the revenue opportunities presented by AI models and services. If Meta successfully monetizes its AI infrastructure, it could signal a bullish trend for the entire AI investment landscape.

Conclusion

If Meta can prove its ability to monetize its AI infrastructure, it would not only benefit the company but also positively impact the broader AI investment theme. Current valuations of Meta and other hyperscalers may be significantly underestimated from a long-term perspective.