The artificial intelligence boom is not just making Nvidia rich; it is fundamentally reshaping which companies investors believe will dominate the next decade. A comprehensive analysis of the world's 97 mega-cap companies reveals that while Nvidia remains the standout winner, Broadcom's explosive 586% growth over the past year signals that the AI infrastructure race is rewarding an entire ecosystem of suppliers, from foundries to equipment makers.

Which Chip Companies Are Winning the AI Race?

Nvidia's dominance is undeniable. The graphics processing unit (GPU) designer has increased its market capitalization by 740% since 2023, rising from $598 billion to $5.02 trillion, making it the world's most valuable company. But the real story lies in what happened to the companies that supply the infrastructure Nvidia's chips depend on.

Broadcom, which manufactures the networking and connectivity semiconductors that link AI data centers together, has become the second-fastest-growing mega-cap company. Its market value jumped from $261 billion to $1.79 trillion in the same period, a gain of 586%. This surge reflects investor recognition that AI data centers require far more than just powerful processors; they need the entire ecosystem of components that make those processors useful.

Taiwan Semiconductor Manufacturing Company (TSMC), the world's leading contract chipmaker, has also benefited enormously. The company's market capitalization climbed 379% to reach $2.21 trillion. In June 2026, TSMC posted sales of 443 billion new Taiwan dollars, or approximately $13.99 billion, beating analyst expectations and reflecting continued robust demand for advanced chip manufacturing. The company's second-quarter sales rose 36% year-over-year, driven largely by AI-related demand.

Samsung and ASML, the Dutch company that manufactures the specialized equipment needed to produce advanced semiconductors, have also emerged as major winners. Samsung's market value has nearly quintupled since 2023, climbing from $305 billion to $1.5 trillion, with much of the recent surge driven by soaring demand for its high-bandwidth memory (HBM) chips used in AI accelerators. ASML's market capitalization has increased 178% to $695 billion, reflecting the critical bottleneck its lithography equipment represents in the race to manufacture advanced chips.

Why Are Semiconductor Companies Dominating Market Growth?

The concentration of wealth creation in the semiconductor sector is striking. Four of the eight fastest-growing mega-cap companies since 2023 come from the chip industry: Nvidia, Broadcom, TSMC, and ASML. This dominance reflects a fundamental truth about the AI infrastructure boom: it requires not just innovation in chip design, but also manufacturing capacity, connectivity solutions, and the equipment to produce those components at scale.

The AI boom has extended beyond pure chip designers to reward companies across the entire supply chain. Alphabet, which is not primarily a chip company but has benefited from AI adoption across its cloud and search businesses, has nearly quadrupled in market value since 2023, rising 279% to $4.53 trillion. This suggests that investors see AI as a transformative force affecting multiple layers of the technology ecosystem.

However, not all mega-cap technology companies have benefited equally. Meta, despite more than tripling in value since 2023, has actually declined 9.68% over the past year, suggesting that investor enthusiasm for AI-focused mega-caps has cooled relative to other opportunities.

How to Evaluate Semiconductor Investments in the AI Era

For investors trying to navigate the semiconductor sector, the choice between different investment approaches has become increasingly important. Two major semiconductor exchange-traded funds (ETFs) illustrate the strategic decision facing market participants:

• Concentrated Mega-Cap Approach: The VanEck Semiconductor ETF (SMH) concentrates its holdings in the largest companies, with its top five positions being AMD at 10.33%, Broadcom at 9.57%, Micron at 9.39%, TSMC at 8.75%, and Nvidia at 8.40%. This approach has returned 376% over five years, significantly outperforming equal-weight alternatives.
• Diversified Equal-Weight Approach: The SPDR S&P Semiconductor ETF (XSD) spreads investments nearly evenly across roughly 40 U.S. semiconductor companies, capping individual positions near 3% each. This approach has returned 197% over five years but has outperformed the concentrated approach in 2026, gaining 71% compared to SMH's 65%.
• Geographic and Supply Chain Exposure: SMH includes direct exposure to TSMC and ASML, foreign companies that XSD's U.S.-only mandate completely excludes. This matters significantly because TSMC controls the majority of advanced chip manufacturing capacity, while ASML is the sole supplier of the extreme ultraviolet lithography equipment needed to produce cutting-edge semiconductors.

The performance divergence between these approaches reflects a fundamental question about the AI boom's trajectory. SMH's concentration strategy assumes that mega-cap winners will continue to dominate, allowing investors to benefit from the compounding effect of letting winners run without aggressive rebalancing. XSD's equal-weight approach bets that the AI infrastructure rally will broaden, rewarding smaller and mid-cap semiconductor companies focused on analog, power management, and specialty applications.

What Does Sustained Growth Require?

As valuations climb to record levels, the semiconductor industry faces a critical inflection point. Alan Goldberg, lead data analyst at BestBrokers, explained the shift in investor expectations: "Over the past three years, investors have poured unprecedented amounts of capital into the companies building the infrastructure behind the AI revolution, from chip designers and manufacturers to cloud platforms and digital ecosystems. Yet history suggests that markets rarely move in a straight line. As valuations climb to record levels, expectations inevitably rise alongside them, leaving even the world's most valuable companies under increasing pressure to prove that today's optimism can translate into tomorrow's earnings."

"The next chapter of the AI boom will not be shaped by excitement over potential but by evidence of sustainable, long-term value creation. In that environment, market leadership is likely to become more fluid, with fortunes determined not only by technological breakthroughs but by the ability to consistently exceed exceptionally high expectations," Goldberg added.

Alan Goldberg, Lead Data Analyst at BestBrokers

TSMC's recent performance illustrates this dynamic. The company beat sales expectations in June 2026, yet its stock fell 2.9% following the announcement. This muted market reaction to record financial results suggests that investors are increasingly focused on whether companies can sustain growth rather than simply achieve it. Wedbush Securities analyst Matt Bryson noted that TSMC's strong sales indicate "demand for advanced node foundry production remains extremely healthy due to continued robust AI requirements," and reported that TSMC has been raising prices by 5% to 10% for chips produced on its most advanced process nodes.

The semiconductor industry's role in the AI infrastructure boom is undeniable, but the next phase of growth will depend on whether companies can translate record demand into sustainable profitability. For investors, this means the era where scale alone defined corporate leadership has given way to one where innovation, execution, and the ability to exceed increasingly high expectations determine market success.