NVIDIA’s $600 Billion Wipeout, Is the AI King Losing Its Edge?

The $2.8 trillion chip juggernaut NVIDIA Corporation (NASDAQ:NVDA) announced Q4 earnings, and clearly disappointed the Street. There has been a lot of noise around stock based as well as the threat of generative AI technology from China – DeepSeek in particular – which has put the share price under serious pressure.

That is in stark contrast to the past ten years that resulted in an absolutely extraordinary 22,000% gain. That is a humongous return, especially when considering how the overall market has been tracking over the same time period.

But over the past three months, though, NVIDIA’s stock has clocked in a close to 6% decline and yet, is still trading at a premium valuation given that it sits at 29.61x its forward non-GAAP earnings, higher than the industry average, but lower than its own five-year average.

Is this the time to scoop up NVIDIA shares or is the chip maker too expensive?

Key Point

• NVIDIA’s Q4 revenue soared 78% to $39.33B, beating expectations, yet the stock slumped due to concerns over China’s DeepSeek AI competition.

• NVIDIA holds a 70%-95% market share in AI chips, but DeepSeek’s cost-efficient model challenges its pricing power and future monopoly.

• CEO Jensen Huang expects AI demand to surge 100x, supporting long-term growth.

How Did NVIDIA Get To Be So Big?

NVIDIA blasted onto the investing world in recent years but 30 years ago when Jensen Huang founded the firm his ambitions for the firm were much more modest.

Back then the company kicked off as a modest chip designer making 3D graphics possible in gaming and struck gold when it invented the GPU, or graphics processing unit, that quite literally changed what was possible in computing.

GPUs are substantially quicker for data intensive apps like gaming than CPUs, or central processing units, and so Jensen Huang made this NVIDIA’s focus. A game-changer for NVIDIA took place in 2006 when it released the programming language called CUDA that made parallel processing possible.

Another key ingredient that was transformative in the Jensen’s success came when the investing community discovered that GPUs were foundational to AI LLMs. ChatGPT, the generative AI large language model that took the world by storm, is trained and runs on thousands of NVIDIA GPUs.

Its Hopper Superchips are very popular on the market, and the company’s new Blackwell platform is also seemingly gaining traction. Last year, Mizuho Securities estimated that NVIDIA commands between 70% and 95% of the market for AI chips used for training and deploying models like OpenAI’s GPT. That alone meant NVIDIA went to market with massive pricing power, charging over $30,000 for its most prized chips.

But competitors haven’t rested on their laurels and are running hot on the company’s heels. Then a bombshell hit when NVIDIA faced some pressures from the launch of a low-cost AI model from China by the company DeepSeek.

The Chinese firm gained attention because its V3 AI model required less than $6 million worth of computing power from NVIDIA’s H800 chips. This Chinese AI assistant has surpassed the capabilities of ChatGPT, which required a lot more computing power.

This was seen as concerning for NVIDIA because ChatGPT was the success story that was shown to investors to prove that GPUs are unmatched. This also saw a pledge from big tech for billions of dollars of investment into AI and its development. DeepSeek’s success story put this requirement on shaky grounds.

DeepSeek’s rise led to NVIDIA shedding close to $600 billion in market value, which earned the unenviable record of the largest one-day loss for a U.S. company. On the other hand, the company has remained unfazed when it comes to the push for its Blackwell platform (although this is a costly ramp-up). The Blackwell rollout was expected to put some pressure on margins. On the bright side, big tech is going full blast with its expensive data center spending.

Is NVIDIA Stock Too Expensive?

Despite its enormous run up in share price in recent years, NVIDIA stock is not too expensive according to analysts who calculate 53.4% upside to $172.86 per share.

Following the disappointment of management’s recently reported fourth quarterly and fiscal year results for 2025 the reality is the top line of $39.33 billion was seriously impressive, not least the fact that it grew by 78% from the prior year’s period.

This figure was higher than what the analysts polled by LSEG were expecting of $38.05 billion. Data center revenue came in at a record $35.60 billion, up 93% from a year ago, a clear indication that demand for chips doesn’t seem to be waning.

Prior to the report there was some worry that margin pressure would be evident but that all ended up being false. NVIDIA’s operating income on a non-GAAP basis was up by 73% from the year-ago value to reach $25.52 billion. Non-GAAP earnings per share were $0.89, up by 71%, while analysts were expecting $0.84. As the company has become larger, growth has slowed a bit, but the overall growth figures remain quite robust.

Furthermore, NVIDIA’s CEO, Jensen Huang, has made it known that the next generation of AI models will require 100 times more computing power than older models. He also shrugged off concerns regarding DeepSeek’s success.

So, at this point, the biggest beneficiary of the AI boom is still benefitting. The company has grown a lot, so it would not be wrong to expect that its growth will slow at some point, but at this point, the company’s growth is holding up well.