Trump's AI Chip Export Controls Stall: These 5 US AI Infrastructure Stocks Gain Breathing Room
The Trump administration's aggressive bid to tighten export controls on advanced AI chips has hit significant bureaucratic roadblocks across government agencies, delaying implementation of new restrictions that would have further limited sales to China and other markets. This unexpected slowdown, reported in recent weeks, comes as U.S. companies race to build out domestic AI infrastructure amid surging demand for data centers and compute power. Far from a setback for national security hawks, the delay hands a timely tailwind to American AI hardware leaders, allowing them to capture more share in the exploding U.S. market before any new rules take effect.
Over the past year, U.S. national AI policy has crystallized around bolstering domestic capabilities, with executive orders and funding initiatives prioritizing onshoring of AI compute. NVIDIA's SEC filings highlight past restrictions costing billions—like a $4.5 billion charge on H20 inventory in Q1 FY2026—but the bureaucratic snag means no immediate escalation. This preserves revenue streams and lets firms like NVDA, AMD, and SMCI focus on hyperscaler deals with Microsoft Azure and Meta. With U.S. AI capex projected to hit $200 billion in 2026, the policy framework favors pure-play infrastructure providers over those exposed to international sales.
NVIDIA (NVDA): The AI Compute Kingpin
NVIDIA dominates AI infrastructure with its GPUs powering 90% of data center revenue ($62.3 billion in Q4 FY2026, up 75% YoY). Export control delays are a boon, as China restrictions already foreclosed much of that market; SEC filings note inability to compete there without dual U.S.-China approval. Domestically, Blackwell ramps and sovereign AI (tripled YoY) position NVDA for U.S.-only growth.
| Metric | Value (TTM/FY2026) |
|---|---|
| Market Cap | $4.59T |
| Revenue Growth | 65% |
| EBIT Margin | 60% |
| P/E Ratio | 38x |
| Price Return YTD | -3% |
Verdict: Strong buy. NVDA's ecosystem moat and 60.5% gross margins make it the top beneficiary of policy-driven U.S. AI buildouts.
AMD (AMD): The Value AI Challenger
AMD's MI-series AI accelerators and EPYC CPUs target data center AI, with the segment at 50% of revenue. Like NVDA, AMD faces export hurdles to China but gains from delays allowing focus on U.S. cloud wins (e.g., Azure integration). Data center growth hit 34% TTM amid EPYC adoption.
| Metric | Value (TTM/FY2025) |
|---|---|
| Market Cap | $400B |
| Revenue Growth | 34% |
| EBIT Margin | 11% |
| P/E Ratio | 92x |
| Price Return YTD | -12% |
Verdict: Bullish. At a relative discount to NVDA, AMD offers compelling exposure to diversified U.S. AI chip demand.
Super Micro Computer (SMCI): AI Server Speed Demon
SMCI builds GPU-optimized servers (70%+ revenue from AI platforms), partnering closely with NVDA/AMD. Q2 FY2026 revenue hit $12.7B; full-year guidance upgraded to $40B+. Delays sidestep supply chain risks from export rules, boosting U.S. hyperscaler orders and liquid cooling tech.
| Metric | Value (TTM/FY2025) |
|---|---|
| Market Cap | $15B |
| Revenue Growth | 35% |
| EBIT Margin | 4% |
| P/E Ratio | 17x |
| Price Return YTD | 3% |
Verdict: High-conviction buy. SMCI's low valuation and rapid deployment edge make it a pure-play on U.S. data center ramps.
Arista Networks (ANET): The AI Networking Backbone
ANET's EOS software and 800G switches power AI clusters (AI networking guidance raised to $3.25B for FY2026, up from $2.75B). Cloud titans (48% revenue) like MSFT drive 29% TTM growth. Policy delays ensure U.S.-centric deployments without global export friction.
| Metric | Value (TTM/FY2025) |
|---|---|
| Market Cap | $186B |
| Revenue Growth | 29% |
| EBIT Margin | 43% |
| P/E Ratio | 53x |
| Price Return YTD | 1% |
Verdict: Buy. ANET's 64% margins and AI rack leadership cement its role in domestic infrastructure.
Microsoft (MSFT): The Cloud AI Orchestrator
MSFT's Azure (40% revenue) integrates NVDA/AMD hardware for AI services, with Intelligent Cloud growing 17% TTM. As a policy beneficiary via U.S. gov't contracts, delays enhance its edge over international rivals. Copilot and sovereign AI add tailwinds.
| Metric | Value (TTM/FY2025) |
|---|---|
| Market Cap | $2.75T |
| Revenue Growth | 17% |
| EBIT Margin | 47% |
| P/E Ratio | 23x |
| Price Return YTD | -15% |
Verdict: Accumulate. MSFT's scale and ecosystem make it a stable U.S. AI winner, though less pure-play.
Ranked Conviction: The Policy Tailwinds Hierarchy
- SMCI – Cheapest valuation, fastest growth in AI servers.2. NVDA – Unrivaled dominance, but premium priced.3. AMD – Balanced chip exposure at scale.4. ANET – Essential networking with fat margins.5. MSFT – Safest mega-cap bet.
This bureaucratic breather could extend into 2026, amplifying U.S. AI capex. Risks: Sudden rule finalization (monitor BIS announcements); China retaliation via tariffs; AI hype cooldown if ROI disappoints (watch hyperscaler capex guidance). Signals to track: NVDA Blackwell supply ramps, Q1 FY2027 earnings beats >10%, U.S. AI policy bills in Congress.