Broadcom has projected that its artificial-intelligence semiconductor business could generate “significantly” more than $100 billion in revenue by 2027, signaling the company’s growing ambitions in the rapidly expanding AI infrastructure market. The forecast, shared by CEO Hock Tan following the company’s latest earnings results, underscores how demand from cloud giants and AI developers is transforming the semiconductor industry.
The company’s latest financial results already show strong momentum in the sector. Broadcom reported $8.4 billion in AI-related revenue in its fiscal first quarter, more than double the figure recorded a year earlier. Overall quarterly sales rose 29% YoY to $19.31 billion, beating analysts’ expectations and highlighting the growing contribution of AI infrastructure spending to the company’s business.
Looking ahead, Broadcom expects AI semiconductor revenue to reach about $10.2 billion in the current quarter, suggesting sustained demand from hyperscale cloud providers investing heavily in AI training and inference systems.
The company’s strategy centers on designing custom AI accelerators and networking chips for major technology firms rather than relying solely on off-the-shelf processors. As companies increasingly seek tailored silicon to optimize performance and energy efficiency for their specific AI models, Broadcom believes it is well positioned to capture a growing share of the AI compute market.
Industry analysts note that demand for custom AI chips is rising rapidly as companies attempt to control costs and performance in large-scale AI deployments. This trend is expected to drive significant growth in the broader semiconductor sector, which is projected to surpass $1 trillion in global revenue by the end of the decade, largely due to AI-related spending.
