NVIDIA (NVDA) remains the top conviction play following a massive earnings beat, with a valuation reset making the stock appear "cheaper" at a forward P/E of 22x–25x despite recent price rallies. Investors should look toward the Blackwell architecture ramp and the new Vera CPU as primary growth drivers, with the company transitioning into a "cash cow" via a 25x dividend increase and an $80 billion buyback program. Keep a close watch on the potential SpaceX IPO in June at a $2 trillion valuation, as its emergence as a "Neo-Cloud" provider for AI models could trigger a liquidity event where investors sell other tech holdings to participate. While the "AI Factory" build-out continues, prioritize hardware over software, as names like Intuit (INTU) are being punished despite strong earnings in favor of infrastructure plays. Finally, consider the energy sector as the ultimate bottleneck for AI; while memory makers like Micron (MU) are currently profitable, long-term margins may face pressure from software efficiencies.

NVIDIA (NVDA)

• Financial Performance: NVIDIA reported a massive Q1 2026 beat with revenue of $81.6 billion (up 85% YoY) vs. $79 billion expected. EPS came in at $1.87 (up 131% YoY) vs. $1.77 expected. • Data Center Dominance: Data center revenue hit a record $75.2 billion, driven by the rapid ramp of the Blackwell architecture. • Guidance: Q2 revenue guidance is set at $91 billion (+/- 2%), signaling nearly 100% year-over-year growth. • Shareholder Returns: The company significantly increased its quarterly dividend from $0.01 to $0.25 per share and authorized an additional $80 billion in share repurchases. • New Reporting Segments: NVIDIA is moving to a new framework to better reflect its business: * Data Center: Split into Hyperscale (Cloud giants) and ACIE (AI Clouds, Industrial, Enterprise, and Sovereign AI). * Edge Computing: Includes PCs, Robotics, and Automotive. • CPU Strategy: Jensen Huang highlighted the Vera CPU as a major new growth driver, targeting a $200 billion TAM (Total Addressable Market) for "Agentic AI" (AI that performs tasks autonomously). • China Impact: Revenue from China-based data center products remained at zero due to export restrictions, though the company is monitoring potential future license approvals.

Takeaways

• Valuation Reset: Despite the stock price rally, the massive earnings growth means the Forward P/E ratio is actually compressing (becoming "cheaper" relative to earnings), potentially sitting around 22x–25x. • CPU Market Share: NVIDIA is no longer just a GPU company; it is on track to become a leading CPU supplier, with visibility to $20 billion in CPU revenue this year. • The "Agentic AI" Era: The focus has shifted from simple chatbots to "Agents" that do productive work. This requires massive compute for "reasoning," which sustains long-term demand for NVIDIA's full stack (chips, networking, and software). • Investment Stability: The 25x dividend increase and massive buyback program signal that NVIDIA is transitioning into a highly profitable "cash cow" while maintaining hyper-growth.

SpaceX (Proposed Ticker: SPCX)

• IPO Details: The podcast discussed leaked S1 details ahead of a potential June IPO at a $2 trillion valuation. • Financials: Reported $18.7 billion in revenue for 2025 but remains unprofitable due to heavy CapEx ($20.7 billion). • AI Connection: Anthropic reportedly agreed to pay SpaceX $1.25 billion per month ($15 billion/year) for data center compute through 2029. • Bitcoin Holdings: SpaceX reportedly holds 18,000 BTC at a cost basis of approximately $35,000.

Takeaways

• Neo-Cloud Player: SpaceX is emerging as a massive "Neo-Cloud" provider, leveraging its infrastructure to host AI frontier models like Anthropic. • Market Liquidity: Analysts expressed concern that a $2 trillion SpaceX IPO might cause investors to sell other tech stocks (like NVIDIA) to find the cash to participate.

Investment Themes & Sector Insights

Semiconductor & AI Infrastructure

• Memory Bottlenecks: Discussion focused on Micron (MU) and SK Hynix. While memory is currently a bottleneck allowing for high pricing, there is a risk that "algorithmic efficiencies" (software needing less memory) could eventually "chop down" the high margins of memory makers. • Energy as the Real Bottleneck: Analysts identified power/energy as the ultimate constraint for AI data centers. China currently holds a significant advantage in energy infrastructure (nuclear/solar) compared to the U.S. • Custom Silicon vs. NVIDIA: While Google (TPUs) and Amazon (Trainium) are building their own chips, NVIDIA’s "rentability" and "financeability" remain superior. Banks are more willing to finance data centers using NVIDIA chips because they are seen as "liquid" assets with high resale value.

Software & Consumer Sentiment

• Intuit (INTU): Despite a "triple beat" (Earnings, Revenue, and Guidance), the stock fell 10%. This highlights a "SaaSpocalypse" where the market is currently punishing software names in favor of hardware/infrastructure. • Elf Beauty (ELF): A "double beat" in earnings suggests that while AI is booming, the consumer still has disposable income for retail/beauty, though the company provided a cautious full-year guide.

Takeaways

• Bullish Sentiment: The analysts remain overwhelmingly bullish on the "AI Factory" build-out, noting that we are not in a bubble because the chips are being utilized at 100% capacity immediately upon delivery. • Risk Factors: The primary risks mentioned include energy constraints, geopolitical tensions with China, and potential gross margin dips during the transition to new chip architectures (Blackwell/Rubin).