Nvidia, Broadcom lead 4 AI chip stocks generating $119B in cash

Nvidia Corp., Broadcom Inc., Micron Technology Inc. and Texas Instruments Inc. produced a combined $119 billion in free cash flow over the past year, giving investors a cash-rich anchor as semiconductor stocks swing on AI spending fears.

"The market is rotating toward quality — companies that can fund their own AI buildout without diluting shareholders," said Rachel Kim, semiconductor analyst at Edgen. "Free cash flow is the new battleground metric in chips."

Nvidia alone generated about $119.4 billion in trailing free cash flow, according to Simply Wall St data, while trading at 30.8 times earnings — well below the semiconductor industry average of 58.7 times. A discounted cash flow model estimates Nvidia's intrinsic value at roughly $236 per share, about 14 percent above its current $207 price. Broadcom's cash flow has surged alongside its custom AI chip business with hyperscalers. Micron and Texas Instruments offer exposure to memory and analog chips respectively, segments that generate steady cash returns even as they lag the AI GPU boom.

The cash flow focus comes as the Philadelphia Semiconductor Index swings between AI euphoria and valuation reality. Bank of America's Bubble Risk Indicator hit 0.91 out of 1 for semiconductor stocks in early July, triggering a 6.4 percent single-day drop in the SOXX. TSMC's capital expenditure reset to as much as $64 billion — up from a prior ceiling of $56 billion — has shifted investor attention to cash generation and the cost of staying at the leading edge.

TSMC raised its full-year 2026 revenue growth outlook to slightly above 40 percent but warned that overseas expansion and 2-nanometer ramp costs would dilute gross margins in the second half. Every incremental dollar of capex compresses the free cash flow yields needed to justify the sector's elevated multiples. Nvidia, which designs but does not manufacture its own chips, avoids the foundry capex burden entirely. Its gross margins have consistently exceeded 70 percent, and its Japan AI factory partnership with Toyota, Fanuc, Sony and SoftBank extends its reach beyond cloud data centers into robotics, industrial automation and smart cities — use cases that could sustain demand even if hyperscaler GPU purchases moderate.

Broadcom has carved a parallel path through custom AI accelerators for cloud giants, a business that carries lower margins than Nvidia's GPUs but offers multiyear design-win visibility with locked-in contracts. Texas Instruments operates a manufacturing-heavy model with its own fabs, but its analog chips serve every electronics end market — from automotive to industrial to personal electronics — providing diversification that pure-play AI names lack.

The cash flow thesis faces three threats. Competition from in-house chips — reports that major cloud customers are developing their own AI silicon — could erode Nvidia's and Broadcom's addressable market over time. Tighter US export controls on chip sales to China, highlighted by recent coverage of DeepSeek and China-related risks, may cap revenue growth. And Chinese rival ChangXin Memory Technologies is preparing an $8.55 billion IPO that poses a longer-term competitive threat to Micron, while spot prices for DRAM and NAND have shown signs of softening. Micron shares fell 5.7 percent in a single session after TSMC's capex reset compounded a sector-wide selloff, and the stock now trades at $937 with technical patterns suggesting further downside risk.

For investors weighing AI semiconductor exposure, Nvidia at 30.8 times earnings with $119 billion in trailing free cash flow screens as the most cash-efficient bet but carries the highest expectations for continued GPU dominance. The stock trades about 31 percent below the average analyst target of $302, according to Simply Wall St data, suggesting the market has not fully priced in its cash generation power. Broadcom offers diversification across networking, software and custom chips. Micron and Texas Instruments provide cyclical memory and analog exposure at lower valuations but face more acute competitive and pricing risks. Whether the apparent discount compensates for AI chip competition and export controls is the central question for semiconductor investors heading into the second half of 2026.

This article is for informational purposes only and does not constitute investment advice.