Nvidia, Broadcom, and AMD are behind some of the most in-demand chips of the AI boom—and Semiconductor Manufacturing Company (TSMC) is the foundry that produces them. That relationship is why TSMC's third-quarter 2026 earnings call, scheduled for October 15 at 2:00 a.m. ET (2:00 p.m. Taiwan time), carries weight far beyond Taiwan. Few events on the calendar offer a clearer read on AI chip demand than the report from the company physically making the silicon.

What TSMC Has Already Told the Market

TSMC's own guidance calls for third-quarter revenue of $44.6 billion to $45.8 billion. At the midpoint, that implies a 12% increase from the previous quarter and a 37% rise from a year earlier.

The projection follows a robust second quarter in which TSMC posted $40.2 billion in revenue, up 33% year-over-year. Management is also guiding for gross margins of 65% to 67% in Q3—a measure of what remains of revenue after direct production costs.

Analysts' forecasts largely line up with management's outlook. Consensus estimates sit at $45.8 billion for Q3 revenue—the very top of the company's guided range—while earnings per share are expected to come in around $4.39. With revenue expectations anchored at the ceiling of guidance, the bar is set high, and merely matching it may not be enough to impress.

Why the AI Chip Trio Is Watching Closely

High-performance computing (HPC)—the segment spanning data center processors and AI accelerators—made up 66% of TSMC's wafer revenue in Q2 2026.

All three chip designers depend on TSMC for mass production of their AI-related products. They draw up the blueprints; TSMC's fabrication plants convert those designs into silicon at scale. Because none of the three operates its own chipmaking factories, TSMC's revenue mix also offers a consolidated view of production volumes across the AI chip industry—context no single designer's report can provide.

The Quiet Period and the October Sales Report

Ahead of the call, TSMC enters a designated quiet period from October 5 through October 14, during which it will refrain from engaging with investors—the customary pre-earnings practice of limiting company communications to prevent selective disclosure.

The company does publish monthly sales figures, and its early October release is expected to offer key indicators on how the semiconductor industry is performing. With TSMC largely out of contact during that stretch, the figures stand as the main official checkpoint before the call. The report will not include margins or management commentary, but it can reveal whether revenue is tracking within the guided range.

Raised Full-Year Targets and a Larger Spending Plan

TSMC has lifted its full-year revenue growth outlook to slightly above 40% in US dollar terms, and it has raised its capital expenditure guidance to a range of $60 billion to $64 billion. Capital expenditure covers spending on fabrication plants and production equipment—the manufacturing capacity that foundry customers ultimately draw on for their chips.

What This Means for Investors

Investor attention will likely center on three questions: whether HPC's share of revenue holds at or above the 66% recorded in Q2; whether gross margins land within, above, or below the 65% to 67% range; and what management says about demand heading into the final quarter and beyond.

The capex number also merits attention. A spending range of $60 billion to $64 billion reflects confidence, but it is also a large financial commitment. Any revision to that guidance—in either direction—could indicate how TSMC views the durability of the AI cycle.

A concentration dynamic is also worth noting. TSMC's heavy tilt toward HPC has fueled its growth, yet it binds the company's fortunes to a single theme. When one customer category accounts for two-thirds of wafer revenue, shifts within it ripple through the entire business.

The early October monthly sales report will provide the first clue. The earnings call will supply the rest, including the margin picture and management's outlook.