Nvidia Corp. delivered a surprisingly strong revenue forecast and pushed back on the idea that the AI industry is in a bubble, easing concerns that had spread across the tech sector.
Nvidia Corp. delivered a strong revenue forecast, with sales expected to be about $65 billion in the January quarter, exceeding analysts' average estimate of $62 billion. The company reported a revenue of $57 billion, a 22% increase from the previous quarter and 62% year-over-year. Nvidia said that its net income in the quarter increased 65% to $31.91 billion, up from $19.31 billion in the year-ago period. Both were above analyst estimates.
The outlook signals that demand remains robust for Nvidia’s artificial intelligence accelerators, the pricey and powerful chips used to develop AI models. Nvidia has faced growing fears that the runaway spending on such equipment isn’t sustainable.
“There’s been a lot of talk about an AI bubble,” Nvidia CEO Jensen Huang told on today’s conference call. “We see something very different.” If all it took to calm investor fears about the AI industry was confidence, Huang would have that solved. Nvidia beat earnings expectations and shares gained by 4% in post market trading as it raised its expectations for the current quarter. Chip stocks rose with it, including AMD and Intel. As long as the outlook for Fed rates remains supportive, we could see prices for AI stocks end two weeks of declines and rally at least until Oracle reports on Dec. 9.
One of the biggest problems chipmakers like Nvidia face is managing supply chains in order to keep servicing demand without costs rising so fast they destroy margins, but Huang sounded confident that the company can manage that. It’s not easy, but it is solvable, he said.
But just because Nvidia’s doing well doesn’t mean there isn’t a bubble somewhere. Some tech firms have valuations that bear no relation to their actual businesses. Others are borrowing aggressively to try to elbow their way into a party they were late to. But companies like Nvidia, and the cloud hyperscalers, are racing to keep up with huge demand and in most cases using their own resources, or reasonable leverage, to do so.
Huang argued that investment in new chips from hyperscalers was helping them lower costs. “Hundreds of billions of dollars of Capex is going to have to be invested,” he said. “It’s fully cash-flow funded.”
He also pointed to the other industries and countries that are just starting to deploy AI, in factories or digital biology. He continues to tout agentic AI as the next big thing and defended recent investments in OpenAI saying Nvidia has got a stake in OpenAI and expected that stake to bring “extraordinary returns,” while its deal with Anthropic would mean the Claude model is run on Nvidia chips for the first time.
Nvidia results were an all-round beat. Asset turnover was 35% in the quarter, up from 33% a quarter earlier and return on equity and return on assets both rose. The company expects to keep gross margins in the mid-70s, CFO Colette Kress told analysts. GAAP gross margins were 73.4% in the 3Q and should be 74.8% give or take 50bps, in the 4Q.
As the company expands, inventories are rising fast, as are accounts receivable. Days of inventory climbed to 119 from 78 a year earlier. That appears to be generating downward pressure on free cash flow. Nvidia doesn’t report free cash flow, but cash from operations was $23.8b and capex was $1.6b, so free cash flow would have been about $22.2b, up 32% from a year earlier.
While revenue from China was disappointing, revenue from the US was a huge beat. The company said it got no sizeable orders for H20 chips from China and sold just $50m of the chips (which may also help explain those rising inventories.) Nvidia doesn’t expect any China data center revenue in the 4Q either. Given how much money it’s making in the US, $39b in the quarter, up 164% from a year earlier, the lack of chip sales to China doesn’t appear to be holding Nvidia back.
