-
Nvidia shares fell 12.6% in November amid growing investor concerns about an artificial intelligence bubble.
-
Google’s new cutting-edge models are trained on its own chips, challenging the idea that companies need to spend money on Nvidia’s expensive GPUs.
-
10 stocks we like better than Nvidia ›
November has been a rough month for me. NVIDIA (NASDAQ: NVDA) investor. Despite reporting another blowout quarterly result on Nov. 19, the artificial intelligence (AI) giant’s stock price fell 12.6% from the close of trading on Oct. 31 to the end of trading on Nov. 28.
In November, concerns grew about an artificial intelligence bubble centered on Nvidia. While Nvidia itself makes huge profits selling picks and shovels to AI gold miners, investors want to know just how much gold there is, and whether it’s enough to justify the purchase of Nvidia’s extremely expensive equipment – at least at current levels.
While this concern has been plaguing shareholders for some time, it came to a head this month letterGoogle released its latest artificial intelligence model Gemini 3. The Large Language Model (LLM) is widely regarded as an improvement over recent models from OpenAI and Anthropic, but crucially, the model is trained using Google’s own chips rather than Nvidia’s.
These chips (called TPUs) are specifically designed and optimized for the operations Google uses to train its LL.M., making them cheaper to produce and cheaper to run. This directly challenges the main narrative of Nvidia’s success, that its chips are so much more advanced than its competitors that they deserve the hefty premiums paid for them.
While it’s true that Nvidia’s GPUs are more powerful and flexible than Google’s TPUs, given how capable the Gemini 3 is, it’s reasonable to wonder why Google (or any other large hyperscaler capable of developing its own chips) wouldn’t rely primarily on its own chips.
Yet despite this concern, Nvidia’s third-quarter earnings showed no sign that its orders were slowing — quite the opposite. It again delivered big revenue and profit growth, both year-over-year and quarter-on-quarter, and its gross margins were in line with expectations for a software-as-a-service (SaaS) company rather than a chipmaker.
Nvidia shares have regained their footing, rising about 4.3% so far in December. While peak fears appear to have receded, I think investors should still remain cautious. It’s hard to argue with the eye-popping numbers Nvidia is putting up, but I do question how much longer its centrality to the AI boom, and the AI boom itself, can continue.
