Broadcom AI Revenue Projected to Reach $44B by 2026 as TPU Strategy Gains Momentum

Broadcom Expands AI Packaging Partnership to Address Hardware Bottlenecks

Broadcom has announced an expanded AI packaging partnership with Applied Materials, combining materials engineering expertise with semiconductor and system-design capabilities to accelerate AI product deployment and improve time-to-market1

1

. Under the EPIC platform, Broadcom will access Applied Materials' global research network, including the new EPIC Center in Silicon Valley, to address growing hardware bottlenecks tied to connecting multiple chips within high-performance computing systems1

1

. Futurum analyst Brendan Burke noted that advanced packaging has become one of the biggest constraints on AI computing growth as chipmakers push for higher interconnect density, faster data movement, and better thermal management1

1

. The EPIC platform gives Applied Materials earlier involvement in chip architecture development, potentially helping Broadcom accelerate deployment of future AI infrastructure technologies.

AI Revenue Trajectory Shows Substantial Growth Through 2027

Analysts project Broadcom's AI revenue could reach approximately $44 billion in calendar year 2026, growing to $78.4 billion by calendar year 2027, with some projections extending to $82.7 billion by fiscal year 20282

2

. This ambitious growth trajectory builds on the company's already strong performance, with revenue reaching $68.3 billion in the last twelve months and growing at 25% year-over-year2

2

. The AI chip stock has emerged as a prominent player in the AI semiconductor market, with a market capitalization of $1.95 trillion and an impressive 83% return over the past year2

2

. Wall Street expects Broadcom to report fiscal second-quarter earnings of $2.32 per share on revenue of $22.08 billion, compared with earnings of $1.58 per share and revenue of $15 billion a year earlier1

1

.

TPU Strategy Positions Broadcom as NVIDIA Competitor

A central element of Broadcom's AI strategy involves Tensor Processing Units (TPUs), specialized chips designed for machine learning workloads. Analysts highlight the company's relationship with Google as particularly significant, with projections suggesting TPU shipments could reach 7 million units annually by calendar year 20282

2

. Google's decision to make TPU technology available to third-party customers positions Broadcom as a more direct competitor in the AI accelerator market2

2

. The successful launch of Google's Gemini 3 model, trained entirely on TPUs, demonstrates the technical viability of these chips for demanding AI workloads2

2

. Analysts note that TPUv7 demonstrates competitive advantages in specific workloads, with performance characteristics and cost structure making them attractive for certain types of AI training and inference tasks.

Major Customer Wins Diversify Revenue Base Across Tech Leaders

Broadcom has secured significant commitments from multiple technology leaders, with its customer roster including Google, Meta, Apple, Anthropic, and OpenAI2

2

. Anthropic has emerged as a notable addition, with an $11 billion order expected to contribute approximately $10 billion in revenue by the second half of fiscal year 2026, establishing Anthropic as Broadcom's fourth major customer in the AI ASIC space2

2

. OpenAI represents another significant opportunity, with a 10 gigawatt partnership anticipated to ramp up in late 2026, positioning OpenAI as the company's fifth major customer2

2

. The company's business model centers on designing custom AI chips for large technology firms, alongside its established data center infrastructure product lines, positioning it to capitalize on accelerating demand while maintaining traditional semiconductor revenue streams2

2

.

Analyst Outlook Remains Positive Despite Revenue Estimate Adjustments

UBS raised its Broadcom price forecast to $490 from $475 while maintaining a Buy rating, though the bank lowered its 2026 Anthropic-related revenue estimate to about $8 billion from $21 billion after the partnership shifted from full AI rack deployments to a more traditional ASIC model1

1

. UBS analysts said the revised structure could improve profitability because the new arrangement carries higher margins despite lower revenue volume1

1

. The stock carries a Buy rating with an average price forecast of $479.86, with recent analyst moves including TD Cowen raising its forecast to $500 and Wells Fargo raising its forecast to $5451

1

. The concentration of revenue among a small number of large customers provides visibility into future demand while enabling efficient resource allocation for product development in AI compute and networking technologies.