$META · Parkev Tatevosian, CFA · 2026-10-26

Meta Platforms is developing several ways to leverage its investments in its large language models. We have already seen an increase in engagement among their user base, as their platforms have become more attractive and capable of displaying content that people find more interesting.

Recently, we saw the success of Meta Muse's personal agent, which surpassed 10 million downloads in just over a month and a half.

Therefore, these are important developments for Meta's stock investors, as the company is spending hundreds of billions of dollars in capital expenditures on its AI data centers and capabilities.

Here is a point of evidence from Meta stating that they have seen a doubling of interaction per person with Meta AI since April, when they began rolling out their new models.

But I think there is an opportunity here for wearable devices, especially since Meta started with glasses. There is no dominant player like Apple in that category, and it is something that people can wear, and many already wear, and it can enhance their daily lives as the glasses combined with a personal MetaMuse agent can help you get through your day more effectively.

The strategy is to price these products at a price close to cost in order to expand the scope of use, and then profit from repeat sales through those devices. Let's say Meta Muse has become incredibly effective at helping people as a personal agent, and that you can sell Meta Muse subscriptions built into the glasses for $9.99 a month.

That would be extremely profitable, with high-margin revenues approaching 70% as a gross margin, in my estimation, on offering such a product.

I think many Meta investors, including myself, were pleased to see that they were balancing their investments in virtual reality and Metaverse, and focusing their investments on artificial intelligence, and they now expect the profit and loss for the Reality Labs sector to improve from here.

Therefore, they have reached the peak of losses in their "Reality Labs" sector and expect things to improve from that point onwards. We are still waiting to see when companies like Meta will announce something similar about their capital expenditures.

When will they say, "Okay, this year will be the peak of our AI capital expenditure, and from this point onward it will decline." Isn't that so? Not in 2026. It's unlikely to be in 2027 based on what I've seen from the management team here.

Although they did not provide specific guidance for 2027, they did talk about how they are making large and rapid investments in their artificial intelligence business. Therefore, it doesn't seem to me that they will reach a peak or that spending will decrease.

It seems to me that their spending on capital expenditures will increase next year.

This year they will spend at least 135 billion according to their revised capital expenditure estimates. So I wouldn't be surprised if the figure rose next year, 2027, from 135 to 150, and perhaps even 175 billion in Meta's capital expenditures in 2027.

Now, in order for investors to accept that with open arms, Meta needs to demonstrate continued success like it has recently done. 10 million downloads for Meta Muse in the first month and a half.

This type of success. Increased engagement on Facebook, Meta platforms, and Instagram, in order to generate more revenue through advertising. This type of success could justify massive spending plans on artificial intelligence.