AI Rally Posts Strongest Momentum Since Dot-Com Era as Geopolitical Risks Mount
Tech sector concentration reaches historic 40% of S&P 500 amid Iran oil shocks and semiconductor bottlenecks, testing whether 25% earnings growth can sustain valuations through macro stress.
The Nasdaq gained 14% in a single month through early May as AI-driven equities posted their strongest momentum rally in three decades, with tech accounting for 85% of S&P 500 year-to-date gains despite mounting geopolitical and supply chain risks. The rally has pushed market concentration to historic extremes — the top 10 S&P 500 constituents now represent nearly 40% of the index, the highest level ever recorded and well above the historical average of 23%, according to Northwestern Mutual.
The momentum is backed by exceptional earnings performance. Tech sector earnings growth hit 45% in Q1 2026, per The Motley Fool, while Morgan Stanley estimates the Magnificent Seven will deliver 25% net income growth in 2026 compared to 11% for the rest of the S&P 500. Morgan Stanley strategist Sam Coffin attributed the recalibration to data showing “an even faster acceleration of AI capabilities than we expected,” leading the firm to raise its 12-month S&P 500 target to 8,300 — a 12% upside driven by stronger earnings rather than multiple expansion.
Regional Markets Surge on Chip Demand
Asia-Pacific semiconductor markets have posted the fastest gains of any major region. South Korea’s Kospi index rallied 80% in 2026 — the steepest climb among major markets — as the country’s market capitalisation reached $4.28 trillion, ranking eighth globally, according to Bloomberg. The index surged from 5,000 to 8,000 points in 2026 alone, driven by memory chip producers supplying AI infrastructure. Taiwan’s Taiex rose 40%, with TSMC projecting the global chip market will reach $1.5 trillion by 2030, with AI and high-performance computing accounting for 55% of demand.
Memory stocks have led the charge. Micron gained 154% year-to-date as of mid-May, while SanDisk surged 493%, according to U.S. News & World Report. Seagate’s fiscal Q3 2026 revenue rose 44% year-over-year to $3.1 billion, while Western Digital posted a 45% increase to $3.3 billion. Memory market revenue is expected to triple in 2026 to $633 billion, driven by AI inference workloads that create persistent storage demand beyond traditional commodity cycles.
Energy Shocks Collide with Chip Supply Constraints
The rally is unfolding against severe macro headwinds. WTI crude peaked at $94 per barrel in April-May — a 57% surge from $60 in late January — as Iran conflict escalated, with one scenario projecting $110 per barrel in Q2. The Federal Reserve Bank of Dallas characterised the Iran situation as the largest geopolitical oil supply disruption in history, threatening 20% of global oil supplies via Strait of Hormuz closures — two to three times larger than the 1973 or 1990 crises. A three-quarter closure would increase headline PCE inflation by 1.1 percentage points, according to Dallas Fed modelling.
US producer prices surged in April, with PPI climbing 1.4% monthly and 6% annually, while petrol prices jumped 15.6%, data from Moneybase showed. The energy shock coincides with semiconductor supply chain fragmentation. Memory chip shortages are projected to drive 50% price spikes by mid-year, while TSMC is implementing foundry price increases of 3-10% at sub-5nm nodes, per Sourceability. ASML is expanding extreme ultraviolet lithography production to 60+ units in 2026 and 80+ in 2027, up from a historical 40-50, to meet surging demand.
“Valuations are no longer uniformly cheap, particularly in the most crowded semiconductor names, but they remain reasonable relative to the strength of earnings revisions.”
— Clarence Li, Lead Portfolio Analyst, T. Rowe Price
Capital Concentration Raises Momentum Reversal Risk
The structural shift in capital allocation is unprecedented. AI companies captured 67% of all private market dollars in 2025 — up from 9% in 2022 — with $94.6 billion flowing to AI ventures, according to Forge Global. AI startup valuations are trading at 10x–50x revenue multiples in 2026, with a median around 20x–30x, reflecting expectations for exponential revenue scaling that have yet to materialise at scale.
Goldman Sachs strategist Ben Snider warned that similar momentum rallies since 1980 have historically preceded weak returns in subsequent months, per IndexBox. The concentration limits diversification opportunities — Goldman noted that with tech accounting for 85% of S&P 500 gains, active stock selection becomes increasingly constrained. Taiwan’s Taiex now trades at 18.2 times forward earnings, above its five-year average of 16x, while Korea’s Kospi trades at 7.6x compared to a five-year average of 10.2x, suggesting valuation divergence between momentum and historical norms.
| Metric | Mag 7 | S&P 493 |
|---|---|---|
| Earnings growth | 18% | 11% |
| Net income growth | 25% | 11% |
| Share of S&P 500 YTD gains | 85% | 15% |
Broadcom reported AI chip revenue of $8.4 billion in Q1 fiscal 2026 (ended February 1), with analysts expecting bottom-line earnings to double within two years. Cisco posted record Q3 revenue of $15.8 billion driven by AI networking demand, triggering a 20% after-hours stock surge. Alibaba’s cloud revenue rose 38% in the latest quarter, with AI-related revenue maintaining triple-digit growth for an eleventh consecutive quarter.
What to Watch
The durability of AI valuations faces three near-term stress tests. First, any Strait of Hormuz closure extending beyond one quarter would push headline inflation beyond Federal Reserve tolerance, potentially forcing credit tightening that would constrain the capex cycles underpinning AI infrastructure buildouts. Second, semiconductor supply constraints at sub-5nm nodes create execution risk for hyperscalers deploying next-generation training clusters — delays in chip delivery could compress the revenue timelines supporting current earnings multiples. Third, the gap between Magnificent Seven earnings growth (25%) and valuation expansion (trading at 18-50x revenue for AI-pure plays) suggests momentum is outpacing fundamentals, raising the probability of sentiment-driven reversals if quarterly guidance disappoints.
Market breadth indicators will signal whether the rally can broaden beyond the top 10 holdings. If energy costs stabilise below $100 per barrel and semiconductor supply chains absorb current price increases without cascading delays, the 12% upside in Morgan Stanley’s S&P 500 target remains achievable. However, synchronized shocks across energy, chips, and credit would test whether 2026’s momentum can survive the transition from hype-driven capital allocation to cash-flow validation. Taiwan’s forward P/E premium and Korea’s rapid rally have historically preceded mean reversion when macro conditions tighten — the question is whether AI’s structural demand proves resilient or whether this rally follows the playbook of prior momentum-driven advances that Goldman warns have led to weak subsequent returns.