SpaceX Kicks Off High-Grade Bond Sale | Bloomberg Tech 6/22/2026

Key Concepts

• AI Infrastructure & Inference: The shift from training large models to running "inference" (executing models for tasks), which is becoming a massive market.
• Hyperscalers: Large-scale cloud providers (e.g., Microsoft, Alphabet, Amazon) building massive data centers.
• Open-Source AI: Models (like Llama 3 or GLM) that allow enterprises to build proprietary, cost-effective solutions without relying solely on "closed" frontier labs.
• Capex (Capital Expenditure): The massive spending by tech companies on hardware, data centers, and energy to support AI growth.
• Investment-Grade Bonds: Corporate debt issued by companies (like SpaceX) to fund operations, signaling financial stability to institutional investors.
• Memory Chip Cycle: The historical "boom and bust" pattern of semiconductor memory, now potentially being smoothed out by long-term supply agreements.

---

1. SpaceX: Inaugural Bond Sale and Market Strategy

• The Move: SpaceX is entering the corporate debt market for the first time, aiming to raise approximately $20 billion.
• Rationale: Despite holding over $100 billion in cash, the company is raising capital to fund massive future capital expenditures (capex) as it transitions into a "hyperscaler in the sky."
• Market Reception: Bloomberg Intelligence analyst Robert Schiffman notes that because SpaceX is a new name in the bond market, demand is expected to be high (potentially exceeding $100 billion in interest).
• Financial Discipline: SpaceX is committed to maintaining low leverage (below 3x) and keeping $25 billion in cash on hand, mirroring the growth strategies of companies like Amazon and Alphabet.
• Stock Performance: Shares have faced pressure, down for three consecutive sessions following their June 12th IPO, trading around $167 (down from initial highs).

2. AI Infrastructure and Strategic Partnerships

• Micron & Anthropic: Micron has entered a strategic agreement with Anthropic to collaborate on memory and storage architecture for AI. This is a significant shift, as memory makers are now dealing directly with "Frontier Labs" rather than just through hyperscaler intermediaries.
• Base 10’s $1.5 Billion Raise: The inference startup raised $1.5 billion across two tranches (at $11B and $13B valuations). CEO Tuan Hoang and investor Apoorv Agrawal (Altimeter Capital) emphasized that the "app layer" of AI is surging, requiring massive compute procurement and specialized software to make open-source models perform at frontier levels.
• Diversification: Base 10 is diversifying its compute sources across 18 different clouds and 90 clusters to mitigate supply constraints.

3. Energy and Big Tech

• Project Kilby: Microsoft signed a 20-year deal with Chevron to build a massive natural gas-fired power plant in West Texas to support a new data center.
• The "Energy Gap": Tech companies are increasingly turning to fossil fuels (natural gas) to meet the sheer scale of AI power demand, often diluting previous 100% renewable energy commitments.
• Economic Advantage: The project leverages the abundance of "stranded" natural gas in the Permian Basin, which is often burned off or underutilized due to pipeline capacity limits.

4. Corporate Shifts and Market Trends

• Meta/WhatsApp: Meta is investing $900 million in the Indian fintech startup CRED and appointing its founder, Kunal Shah, as the new global head of WhatsApp to drive monetization in the Indian market.
• Apple Design: Incoming CEO John Turnis is reportedly planning a major shakeup of Apple’s industrial design studio, aiming to restore the creative influence the team held during the Steve Jobs/Jony Ive era.
• Semiconductor Outlook: The Philadelphia Semiconductor Index (SOX) is up 90% year-to-date. Analysts expect this momentum to continue for several quarters as earnings estimates for AI-related infrastructure continue to be revised upward.

5. Notable Quotes

• Robert Schiffman (Bloomberg Intelligence): "I think three-quarters of the people that are going to buy the bonds aren't even going to know what this company [SpaceX] does. It's going to be very much of a buy now, ask questions later."
• Apoorv Agrawal (Altimeter Capital): "Inference is going to be one of the largest, if not the largest, markets in the world... it's not just cost, it's capability, control, and cost."
• Nouriel Roubini (via Bloomberg TV): "People are obsessed with the Fed... the key story is the tech boom, and that's going to be the most important first-order impact."

---

Synthesis/Conclusion

The market is currently defined by a massive, sustained "spending party" on AI infrastructure. While equity markets show volatility (as seen with SpaceX and Alphabet), the credit and bond markets are signaling strong confidence in the long-term capex cycle. The industry is shifting from a focus on "frontier model training" to "inference efficiency," where companies like Base 10 and Micron are positioning themselves to provide the necessary hardware and software to make AI profitable at scale. Despite concerns over interest rates and energy emissions, the dominant narrative remains the rapid, multi-year buildout of AI-ready data centers and the integration of AI into every layer of the enterprise.