3 Reasons I Think Meta Platforms is a Screaming Buy Right Now

It's been a tough year for Meta Platforms (NASDAQ: META) shareholders. The social media and artificial intelligence (AI) specialist has lagged the broader market, with the stock down 9% over the past year (as of market close on Thursday), compared to 21% gains for the S&P 500.

The biggest headwind has been fears about Meta's significant AI-related spending, as investors fear the costs will squeeze the company's profits and ultimately outweigh the benefits.

However, I believe the sell-off has gone too far. Numerous catalysts could send Meta stock soaring over the past year, and I don't think investors have been keeping track. Let's look at these drivers and why I think Meta Platforms is a screaming buy right now.

Image source: The Motley Fool.

1. Head in the clouds

The tech world was rocked last week when rumors surfaced that Meta is developing a cloud infrastructure business, according to a Bloomberg report. This would put the company in direct competition with industry leaders Amazon Web Services (AWS), Microsoft Azure, and Alphabet's Google Cloud.

One of the key advantages for cloud operators, aside from selling on-demand computing capacity, has been the built-in market for selling AI models. Amazon Bedrock, Google Cloud Vertex AI, and Azure AI Foundry give customers access to top-tier AI models and services without the need to spend heavily on underlying infrastructure. This has given the Big Three cloud providers a significant advantage over Meta. If the company develops its own cloud service, that would level the playing field.

Meta has been investing heavily in data centers to support its AI ambitions. Having an outlet to sell excess cloud capacity and peddle its homegrown models would be a boon to Meta and take its AI strategy to the next level.

2. Significantly lower AI infrastructure costs

One of the biggest question marks hanging over Meta this year is the company's aggressive AI-related capex spending. In the first quarter, Meta raised its forecast, saying it expects spending to be in a range of $125 billion to $145 billion, up from its previous range of $115 billion to $135 billion. Much of this spending will advance its data center build-out to support its AI ambitions. Some investors fear Meta's spending will outpace the returns from the company's massive investment.

However, Meta's AI infrastructure build-out is much more cost-effective than expected, according to BofA analyst Justin Pope. The company is working to deliver an additional 14 gigawatts (GW) by the end of next year, with 1GW already online. The analyst originally estimated Meta's cost per GW at $45 billion, but an internal Meta memo suggests the cost is closer to $22 billion. If those figures are "even close to accurate, Meta may have engineered significant cost savings" that are well below Wall Street's expectations, according to the analyst.

Moreover, the company has joined forces with Broadcom to develop a suite of Meta Training and Inference Accelerator (MTIA) chips. These specialized processors would be designed to be more efficient for specific tasks, thereby reducing AI-related operating costs. The first of these custom chips, dubbed "Iris," is scheduled to begin production in September, according to reports.

3. Muse Spark reception

After the tepid reception to its Llama 4 AI model last year, Meta took a step back to regroup. Just this week, the company released Muse Spark 1.1, and early reviews suggest the company has a hit on its hands. The latest multimodal AI -- which powers the Meta AI assistant -- offers advanced reasoning and can handle complex processes. The recent update also offers agentic coding, or the ability to write, test, and debug code with minimal human involvement.

While these latest models still lag those from OpenAI and Anthropic, Meta has significantly narrowed the lead with Muse Spark. Moreover, the cost of use is lower than that of the leaders, as CEO Mark Zuckerberg has promised "aggressive" pricing and a "much more affordable cost" for its frontier model. Developers testing Muse Spark will be able to use it for free, though they will be forced to pay beyond a certain use threshold.

The creation of this next-generation AI model catapults Meta into the big leagues, and its aggressive pricing will no doubt attract serious users.

A screaming buy?

Despite these significant developments and its expanding opportunity, Meta still trades at a discount to many of its Magnificent Seven rivals. The stock is currently selling for less than 23 times earnings (as of market close on Thursday).

That's an attractive price for a company with so many ways to win, which is why I believe Meta Platforms is a screaming buy right now.

Bank of America is an advertising partner of Motley Fool Money. Danny Vena, CPA has positions in Alphabet, Amazon, Broadcom, Meta Platforms, and Microsoft. The Motley Fool has positions in and recommends Alphabet, Amazon, Broadcom, Meta Platforms, and Microsoft. The Motley Fool has a disclosure policy.