Why Nvidia Is the Last AI Stock I’d Buy Right Now

Written By

Dan Buckley

Updated

Aug 28, 2025

Yes, Nvidia is the largest AI public company in AI right, and yes, it’s the stock on everyone’s lips, with over 20 million people a month searching for ‘Nvidia stock’.

However, that’s also why I would be extremely cautious about buying it currently…

Quarterly Results: Good But Still Room for Improvement

Nvidia’s Q2 results delivered $46.7 billion in revenue (up 56% year-over-year) and $1.05 in adjusted earnings per share, comfortably beating most expectations. The company also launched a $60 billion share buyback and forecasted Q3 revenue of around $54 billion.

Still, we saw shares dip about 3% after hours due to weaker-than-expected data center sales and lingering uncertainty over China’s chip demand, especially its H20 shipments.

Valuation Metrics Are Flashing Warning Signs

The company has a $4.4 trillion valuation, making it the first US firm to hit the $4 trillion mark. But while it’s easy to get caught up in hype and headlines, this run-up creates forward-looking risks:

Its forward P/E ratio sits at around 33.7x, meaning investors are willing to pay about 34 times expected future earnings – a sign the market is banking on strong profit growth.
By contrast, its trailing P/E is in the high 50s, which shows that based on the past 12 months of actual earnings, the stock already looks very expensive.

Dominance Doesn’t Guarantee Hypergrowth

Yes, proponents will tell you Nvidia still dominates the space, and to be fair, it does control roughly ~92 % share of discrete GPUs and over 80% of AI chips used in training and deployment. But it’s also starting to mature into its market.

Most analysts now expect 53% year-on-year revenue growth this quarter, or about $46 billion. Still big, okay gigantic numbers, but let’s not forget it’s down from the eye-popping 122% growth we saw a year ago.

What I’m saying is – the hypergrowth days are slowing. Therefore, it’s more likely than not that Nvidia will produce more normal growth in its stock price over the next 5-10 years than the massive 10x+ gains that we’ve seen. There’s simply only so much market capacity, even when AI is involved.

A Reality Check on Market Capacity

As a basic rule of thumb, you can take a company’s market cap and divide it by 10. Can it make that much in profits in 5 years’ time? Can Nvidia really make $400+ billion per year in annual profits by 2030?

It’s not something to take overly literally, but I find it can help frame what a valuation might mean in more concrete terms.

AI Hype vs. Real Adoption

One issue is that Nvidia’s sky-high valuation assumes the AI boom keeps running at full throttle. We just don’t know yet.

One MIT study found that 95% of AI pilots fail. 95%, that’s staggering. Regardless of the actual figure, because it’s a fluid situation, it’s something you’ve got to keep in mind when thinking about how much in future corporate spending Nvidia’s stock price depends on.

Most of us have also seen that OpenAI’s CEO has warned that investors are getting carried away with AI hype and that fuzzy terms like “AGI” (Artificial General Intelligence) are losing relevance.

Of course, Nvidia might still keep climbing if its financials continue to match or beat expectations, long-dated projects are continually discounted into its valuation, and the overall macro environment remains good for the stock market.

As we’ve seen with companies like Tesla, if you can keep investors focused on far-out projects, that can sustain a valuation for a long time. All you have to do is buy a social media behemoth and cosy up to the world’s most powerful politician, easy, right?

The Risks Are Piling Up

Markets are ultimately discounting mechanisms. In equities, that means years, even decades, of future earnings can be priced into a stock. But don’t forget the risks are there.

Export restrictions have already cost Nvidia $4.5 billion in lost sales to China. I’ve seen estimates that are up to $8 billion in potential impact next quarter.

So if spending cools off or AI monetization proves weaker than promised (and it’s got seriously big boots to fill), then generating the hundreds of billions of dollars in profits to justify the current valuation becomes increasingly difficult.

And Nvidia doesn’t need to see a fall in revenue or earnings for the stock to re-rate, just a less rapid rise relative to what’s discounted in. Think about it like that, and Nvidia starts to look a lot less like a “sure thing”, at least in my view.

Bottom Line

When a stock starts trading at nearly 60x established earnings like this, even small setbacks can trigger painful sell-offs because so much of the stock price is in future expected performance.

Because Nvidia is a longer-duration growth story, any broader market dip will also tend to hit it disproportionately. For all its current dominance in AI, this is one name where I’d be very careful about going all-in.

Concentrated AI bets may give your portfolio a better “right tail” – in other words, if one of those stocks soars, your returns could climb much higher than average.

But that potential reward may not make up for the asymmetric downside risk – meaning if those few stocks fall hard, your losses could be much bigger than the gains, because you don’t have other investments to balance it out.

This article reflects the author’s personal opinion and is provided for informational purposes only. It should not be considered financial advice. Always do your own research or consult a licensed financial advisor before making investment decisions.

Article Sources

The writing and editorial team at DayTrading.com use credible sources to support their work. These include government agencies, white papers, research institutes, and engagement with industry professionals. Content is written free from bias and is fact-checked where appropriate. Learn more about why you can trust DayTrading.com