Broadcom on Wednesday signaled that revenue for its fiscal third quarter will top analysts' expectations, driven by healthy demand for its purpose-built AI chips and networking hardware. The Palo Alto, California-based semiconductor and infrastructure-software firm nevertheless saw its shares decline by more than 2% in after-hours trading, following a nearly 8% gain in the prior week ahead of the earnings release.
The company has emerged as an important vendor in the AI ecosystem, supplying custom processors to hyperscale customers including Meta and Alphabet's Google as an alternative to the dominant Nvidia devices, which are often described as costly and energy intensive. Broadcom emphasized that as machine learning applications evolve, requirements differ substantially across customers, prompting large cloud providers to develop their own processors to reduce costs and tailor workloads.
Broadcom provided a revenue outlook for the third quarter, but the specific expected figure was not clear in the material available; analysts' average estimate stands at $28.54 billion, according to data compiled by LSEG. In a statement on prospects for the next quarter, CEO Hock Tan said: "In Q3 we expect semiconductor revenue from AI to grow over 200 percent year-over-year to $16.0 billion."
For the second quarter, the company reported revenue of $22.19 billion, an increase of 48% year-over-year. That top-line result was slightly below consensus expectations of $22.27 billion. Broadcom also reported adjusted earnings of $2.44 per share, which the company said trailed estimates of $2.40.
The mixed reaction in the equity market - an after-hours decline following recent gains - highlights short-term volatility even when a company positions itself as a beneficiary of AI-related spending. Broadcom's role supplying custom silicon and networking gear to major cloud and internet firms places it at the intersection of semiconductors, cloud infrastructure and enterprise networking markets.
Investors and industry observers will be watching upcoming quarters for continued strength in AI-related semiconductor revenue and for clearer guidance figures from the company. At present, the company points to strong AI semiconductor growth while the published guidance contains some gaps in the specific numerical disclosure available to the public.