Broadcom Just Doubled Its AI Revenue and Sees $100 Billion in Chip Sales Ahead — Is This the New AI King?

The Numbers That Made Wall Street Take Notice

In a week dominated by war headlines and job losses, Broadcom quietly delivered one of the most impressive earnings reports of 2026. The semiconductor giant posted Q1 fiscal 2026 revenue of $19.31 billion, up 29% year-over-year, beating the consensus estimate of $19.21 billion. Adjusted earnings per share came in at $2.05, topping expectations of $2.03. But the number that truly electrified investors was AI infrastructure revenue: a record $8.4 billion, representing a 106% surge compared to the same period last year. Broadcom shares jumped over 5% in after-hours trading, providing a rare bright spot in an otherwise brutal week for markets.

The $100 Billion AI Chip Forecast

CEO Hock Tan made a statement that reverberated across the semiconductor industry: Broadcom has line of sight to over $100 billion in AI chip revenue, with the company expecting to surpass that figure significantly by 2027. This is not a vague aspiration. Broadcom builds custom AI accelerators for three of the largest hyperscalers — widely believed to be Anthropic, Meta, and OpenAI — and its networking chips are essential infrastructure for the massive data centers powering the AI revolution. Q2 guidance came in at $22 billion in revenue, representing an acceleration to 47% year-over-year growth. Adjusted EBITDA hit a record $13.1 billion, representing 68% of revenue — a margin that most companies can only dream of.

Broadcom vs NVIDIA: The AI Chip Rivalry Heats Up

While NVIDIA remains the undisputed leader in GPU-based AI training, Broadcom is carving out a massive niche in custom silicon and networking. NVIDIA reported a record $68 billion quarter but saw its stock drop 14% from post-earnings peaks as investors questioned whether the AI trade has become too crowded. Broadcom, by contrast, is gaining share precisely because hyperscalers want alternatives to NVIDIA dominance. Custom ASICs designed by Broadcom offer better power efficiency and lower total cost of ownership for specific AI workloads. The networking business — switches, routers, and optical components that connect thousands of GPUs in data centers — is equally critical. You cannot run an AI cluster without networking, and Broadcom owns that layer. This is not an either-or story. Both companies will thrive. But Broadcom emergence as a $100 billion AI revenue company fundamentally changes the competitive landscape.

Why This Matters Beyond Tech Stocks

Broadcom earnings matter because they validate that AI infrastructure spending is not slowing down despite the broader economic turbulence. In a week where the jobs report showed 92,000 jobs lost and oil prices surged toward $90, the fact that hyperscalers are still pouring tens of billions into AI data centers tells you something important: the AI buildout is not discretionary spending that gets cut in a downturn. It is strategic infrastructure that companies view as existential. This has implications beyond Broadcom and NVIDIA. It supports the entire AI supply chain — from TSMC and ASML on the manufacturing side to Arista Networks and Vertiv on the infrastructure side. If you believe AI spending will continue regardless of the macro environment, the current market weakness is creating entry points in quality AI names that have been dragged down by broader fear.

Should You Buy Broadcom Here?

Broadcom trades at roughly 25x forward earnings, which is not cheap but is reasonable for a company growing revenue at 29% with 68% EBITDA margins and a clear path to $100 billion in AI revenue. The risk is that the broader market sell-off drags the stock lower regardless of fundamentals — in a risk-off environment, even great companies get sold. The opportunity is that Broadcom is one of the few large-cap tech names where the growth story is actually accelerating, not decelerating. For long-term investors, any pullback toward the $180-$190 range would be an attractive entry point. For those already holding, the earnings report confirms the thesis. This is a hold-and-add-on-weakness situation, not a time to take profits on one of the strongest AI infrastructure plays in the market.