AI Stocks Crashed. Now They’re Roaring Back.

AI Investing: Navigating the Rebound & Future Volatility

Key Concepts:

• AI Buildout: The extensive infrastructure development required to support Artificial Intelligence, including data centers, compute power (GPUs), and power connections.
• NeoCloud Operators: Companies building and renting data center capacity to hyperscalers (large cloud providers).
• Capex: Capital Expenditure – funds used by a company to acquire, upgrade, and maintain physical assets.
• Hyperscalers: Large-scale cloud service providers like Meta, Alphabet (Google), Microsoft, and Amazon.
• Sovereign Wealth Funds: Investment funds owned by national governments.
• Custom Silicon: Chips designed specifically for a company’s needs, rather than using off-the-shelf components. (e.g., Google’s TPUs)
• Technocracy: A system of government where decision-makers are selected based on technical expertise.
• Rare Earths: A set of seventeen chemical elements crucial for many technologies, including AI hardware and national defense.
• Leverage: The use of debt to amplify potential returns (and losses).

I. The AI Sell-Off & Financing Concerns

The recent AI stock sell-off, beginning in October and peaking in December, wasn’t driven by fundamental issues with AI itself, but by concerns surrounding its financing. The AI buildout is a massive undertaking, projected to require $520-530 billion in capital expenditure (Capex) by 2026. This requires robust financing pathways, which experienced stress in late 2023.

Luke Lango identifies three primary financing methods:

1. Cash Financing: Companies like Meta, Alphabet, and Microsoft are utilizing their free cash flow to fund AI data center development. This area remains stable.
2. Debt Financing: Oracle, a key player in debt-financed data center construction, faced difficulties with multiple bond sales, indicating stress in the debt markets. Bond values were decreasing.
3. Private Equity Financing: A $10 billion data center project for OpenAI, initially backed by Blue Owl (a private equity firm), stalled when Blue Owl reportedly withdrew funding, signaling stress in the private equity sector.

II. The Turnaround: OpenAI’s Funding & Sovereign Wealth Fund Involvement

The market’s concerns began to ease with news of OpenAI’s planned venture fundraising round – potentially reaching $100 billion. This is significant because such a large raise necessitates investment from sovereign wealth funds, not just traditional venture capital firms. This influx of capital has downstream effects, intensifying the competition in the AI space.

III. Identifying Beneficiaries: Beyond the AI Developers

Lango emphasizes that the real investment opportunities lie not with AI developers like OpenAI, but with the companies benefiting from the increased spending. He highlights three categories of companies poised for growth:

• NeoCloud Operators: Companies building and leasing data center capacity to hyperscalers.
• Memory Suppliers: Companies providing essential memory components for data centers.
• Rare Earth Material Providers: Companies involved in the supply chain of critical materials needed for AI hardware.

IV. Specific Stock Recommendations & Analysis

Several stocks are highlighted as exhibiting strong rebound potential:

• Coreweave (CRWV): Described as a “highly levered digital landlord,” Coreweave experienced a dramatic rise (from $60 to $200) during the initial AI spending surge, followed by a significant decline (to the mid-$60s) during the sell-off. It’s now rebounding. Its highly leveraged business model is both a risk and a potential reward.
• Micron (MU): A memory supplier for data centers, Micron is historically a cyclical stock with significant price volatility. The company recently reported strong earnings, with next quarter’s guidance projecting growth from 60-70% to 150-170%, before the OpenAI funding news. Lango compares Micron’s earnings beat to Nvidia’s.
• Applied Digital (APLD) & Nebius (NBIS): Smaller NeoCloud operators with faster growth rates than Oracle. They are considered more nimble and less burdened by legacy businesses.
• Oracle (ORCL): While a NeoCloud operator, Oracle is considered less exciting than Applied Digital and Nebius due to its slower growth and existing legacy compute business.

V. The Challenge to Nvidia’s Dominance

Despite its current market leadership, Lango expresses skepticism about Nvidia’s long-term prospects. He argues that Nvidia has a reputation for being arrogant and inflexible in its dealings, prompting major tech companies (Meta, Microsoft, Amazon, Alphabet) to develop their own custom silicon (AI chips).

• Google’s TPUs: Google’s Tensor Processing Units (TPUs) are cited as the most advanced custom AI chip outside of Nvidia’s GPUs. The success of Google’s Gemini 3 model, built on TPUs, prompted Meta to begin purchasing TPUs instead of Nvidia GPUs.
• Amazon & OpenAI Collaboration: A $10 billion deal between OpenAI and Amazon for Amazon’s Tranium chips further demonstrates the shift away from Nvidia.
• Software Competition: Meta and Google are also collaborating on software to compete with Nvidia’s CUDA platform, which is a critical component of Nvidia’s ecosystem.

VI. The Rise of State-Backed Capitalism & Rare Earths

Lango posits that the US is entering a new era of “state-backed capitalism,” where Silicon Valley and Washington D.C. collaborate to prioritize industries crucial for national security and economic competitiveness. This is exemplified by the focus on securing the supply chain of rare earth materials, essential for AI hardware and defense technologies. He recommends reading Alex Karp’s The Technological Republic for a deeper understanding of this trend. MP Materials is highlighted as an early beneficiary of this government support.

VII. Volatility & Political Risks in 2026

Lango anticipates continued volatility in 2026, particularly leading up to the midterm elections. He believes AI will become a significant political issue, with potential regulatory changes depending on the election outcome. He predicts a strong first half of the year, followed by a choppy third and fourth quarter.

Notable Quote:

“The one thing I would watch as a potential wild card is going to be as we get into the midterms in late 26. I think AI becomes a political hot topic.” – Luke Lango

Conclusion:

The recent AI stock rebound is driven by renewed confidence in financing, particularly the massive funding round for OpenAI. Investment opportunities lie not just with AI developers, but with the companies supporting the AI buildout – particularly NeoCloud operators, memory suppliers, and rare earth material providers. While the long-term outlook is positive, investors should anticipate continued volatility, especially in 2026 due to political factors. The dominance of Nvidia is being challenged by major tech companies investing in custom silicon and software solutions.