2 Tumbling Mega-Cap Stocks That Could Keep Crashing


It can be tempting to take advantage of slumping stock prices. Sometimes the market goes overboard, pushing a particular stock into bargain territory. At other times, a big decline is just the beginning as the story that underpins a stock begins to unravel.

Both Nvidia (NASDAQ: NVDA) and Tesla (NASDAQ: TSLA) are getting slammed as investors hit the sell button amid growing unease about the economy, tariffs, and inflation. Here’s why neither stock looks like a buy.

Shares of Nvidia, the overwhelmingly dominant supplier of artificial intelligence (AI) accelerators, have slumped around 20% from their 52-week high. The company’s results remain impressive, and demand for its AI accelerators is still booming. Nvidia’s data center segment generated more than $35 billion in its latest quarter, and profit margins remain sky-high.

Despite these results, market optimism appears to be fading. DeepSeek, a Chinese AI start-up that managed to train a top-tier AI model far more efficiently than equivalents from U.S. companies, upended the notion that more powerful models will require ever-increasing computing horsepower. There are also indications that AI companies are hitting a ceiling on capabilities. Case in point: OpenAI’s latest GPT-4.5 model, which is wildly expensive and represents a minor improvement at best over previous models.

Here’s the fundamental question: Are there enough use cases for AI to justify many hundreds of billions of dollars of spending on AI infrastructure, including Nvidia’s AI chips and everything else that goes into a data center, annually? Increasingly, companies seem to be running into roadblocks actually implementing AI in useful ways. Apple, for example, delayed its revamped Siri assistant because the company is reportedly struggling to deliver on its promised features. OpenAI’s Operator, which uses AI agents to perform tasks for users, was called “brittle and occasionally erratic” in a review by The New York Times.

There are plenty of legitimate use cases for AI technology, but there appears to be a good chance that the industry is greatly overselling the potential of this technology. Ultimately, large language models (LLMs) like those from OpenAI simply predict the next token, like a piece of text or a pixel in an image, in a stream of tokens. That’s all they do. There’s no real reasoning or thinking, and it’s a far cry from how human intelligence works.

Nvidia’s growth story depends on AI becoming increasingly more capable over time. If that’s not happening, the massive build-out of AI computing capacity going on right now could be a one-time event that leads to oversupply. In that scenario, demand for Nvidia’s AI chips would fall off a cliff, and the stock would likely follow suit.



Source link