Artificial intelligence (AI) has undoubtedly been the hottest sector in the stock market over the past few years. Those who saw this trend coming five years ago have likely generated incredible returns. One exchange-traded fund (ETF) that has done well is the Vanguard Information Technology ETF (VGT 0.67%), which is up over 143% in the past five years, and up over a whopping 1,400% since the ETF launched in 2004.

While there are many questions about AI's capabilities, and the sector likely won't move in a straight line similar to the internet boom, there is a high likelihood that AI will significantly change our daily lives and has a long runway ahead. Could a $50,000 investment in VGT turn into a million?

Person looking at charts on large monitor.

Image source: Getty Images.

All kinds of AI exposure

VGT is a passively managed ETF that seeks to give investors broad exposure to the information technology space, which has rapidly become consumed by the largest technology conglomerates that are pouring tens of billions into AI and capital expenditures that will power the AI revolution. Here are the top 10 holdings in VGT by percentage of the fund allocated to each stock (as of Sept. 21):

Company Percentage of the fund
Nvidia 17.18%
Microsoft 13.73%
Apple 13.05%
Broadcom 4.32%
Oracle 2.02%
Palantir 1.86%
Cisco Systems 1.50%
Advanced Micro Devices 1.44%
Salesforce 1.32%
International Business Machines 1.23%

Source: Vanguard.

As you can see from the list above, the largest holdings are Nvidia, the main pick-and-shovel semiconductor play driving the AI ecosystem, and Microsoft, which is one of the main hyperscalers pouring billions into data centers. Microsoft is also developing tons of AI tools that companies can leverage to make their businesses more efficient.

Apple hasn't advanced in AI as much as the other stocks in the Magnificent Seven. However, many still see Apple as a major AI player, especially as it integrates more AI features into its large consumer product set, from iPhones to AirPods.

The list above also includes AI data decision-making company Palantir, whose stock has been invincible this year, and Oracle, which is expected to benefit massively from its AI cloud business. In Oracle's most recent quarter, management said that the company had over $450 billion of revenue performance obligations.

The list above also includes some more legacy internet players, like Cisco and IBM, that stand to benefit from AI but whose stories aren't quite as exciting as the likes of Nvidia, Microsoft, and Palantir.

Can a $50,000 investment in VGT turn into a million?

For a $50,000 investment to turn into $1 million, VGT would have to experience significant appreciation and rise 20x from current levels. Recall that the ETF is up over 1,400% over the last two decades or so, which is certainly nothing to sniff at. Many investors believe AI is still in the early innings, and while they might be right, the largest stocks in this ETF -- Nvidia, Microsoft, and Apple -- have already achieved a certain critical mass.

Nvidia is now the largest stock in the world by market cap, with a nearly $4.3 trillion market cap. Microsoft has a $3.85 trillion market cap, while Apple has a $3.64 trillion market cap. So just to double in value, these stocks have to ascend to enormous market caps. Over time, as companies get bigger, generating the same revenue and earnings growth and commanding the same high multiples becomes more difficult, as their total addressable markets mature. At the end of August, VGT traded at 39 times earnings and had an earnings growth rate of roughly 32%.

That's not to say AI won't continue to climb as it seeps into more sectors and companies large and small, but I think a 20x return from here may be a lofty goal. It will take several decades at the very least. The good news is that the longer you hold a stock or ETF, the less likely it is to lose money. Investors can continue to buy VGT and should be able to generate good long-term returns. However, they should also understand that the road may be bumpy along the way. That's why I would recommend dollar-cost averaging in order to smooth out your cost basis over time.