Meta Platforms remains one of the most undervalued names in big tech, trading at a significant discount despite posting stunning financial results and positioning itself as a leading artificial intelligence model maker. The company's stock trades at just 19 times earnings and 6 times sales, a valuation that appears to ignore both its dominant advertising business generating over 25% annual growth and its emerging AI ambitions.

Why Is Meta Trading So Cheaply Despite Strong Earnings?

The answer lies in investor skepticism about where Meta's massive capital spending will lead. The company is funneling essentially all of its cash flow into building new data centers and AI infrastructure, depressing free cash flow to just $784 million in the most recent quarter. This heavy investment has left the market unwilling to value Meta based on the earnings generated by its core advertising business, which includes Facebook, Instagram, and WhatsApp. Instead, investors appear to be assigning zero value to the company's AI initiatives, treating the capital expenditure as a bet rather than a strategic investment.

The skepticism is understandable given Meta's history. CEO Mark Zuckerberg has spent billions on the metaverse with little to show for it, and the company continues to post hefty losses in its Reality Labs division even as it ramps up AI spending. However, analysts argue this time may be different. The company recently rolled out Muse Code, a direct competitor to Claude Code from Anthropic and Codex from OpenAI, followed by the release of open-source models including Muse Glimmer and announcements to release weights for Muse Spark 1.2.

What Multiple Paths Does Meta Have to Monetize AI?

Meta's AI strategy offers several potential revenue streams that the market currently ignores. The company can pursue different approaches simultaneously, creating what investors call "optionality," or multiple paths to success. This flexibility is a significant asset that doesn't show up in traditional valuation models.

• Frontier Model Tokens: Meta could sell access to its AI models directly to other companies and developers, similar to how OpenAI and Anthropic operate, generating revenue from each query or interaction processed through the system.
• External Compute Sales: The company could sell excess computing power to third parties, allowing other organizations to run their own AI models on Meta's infrastructure, creating a cloud computing revenue stream.
• Internal Product Enhancement: Meta could use AI to improve its core products, including better content recommendation algorithms for its Family of Apps and more sophisticated advertising targeting, directly boosting the company's dominant ad business.

During recent earnings calls, management emphasized that they believe their internal uses for computing power may present higher potential returns than selling excess capacity to external customers. Management also noted their belief that "there's just nowhere near enough compute for all the demand," a statement that reflects the intense competition for AI infrastructure.

The company generated 28% year-over-year revenue growth in its most recent quarter, including 28.2% growth in its Family of Apps businesses. This growth came from increases in both the number of ads delivered and the average price per ad, demonstrating that Meta's core business remains robust even as the company invests heavily in AI.

How Should Investors Think About Meta's Valuation?

To understand Meta's potential upside, consider what its advertising business alone should be worth. Historically, Meta's advertising operations have generated around 50% or greater operating margins. If the company's Family of Apps businesses maintain a 50% net margin over the long term, a conservative assumption, the stock would be priced at around 12 times long-term earnings when excluding Reality Labs losses. Yet the stock trades at just 19 times current earnings, suggesting significant undervaluation even before accounting for any AI monetization.

Management guided the company to see third quarter revenue between $61 billion and $64 billion, representing 19% to 25% year-over-year growth. Consensus estimates call for the company to sustain near 20% top-line growth in the near term, yet the stock trades at just 6 times this year's sales. For comparison, a price-to-earnings growth ratio of 1 times, which is considered fair value for a growing company, would imply a fair value of around 25 times earnings or 12 times sales, assuming 50% long-term net margins.

The market's hesitation appears rooted in uncertainty about management's ability to monetize AI effectively. Management has repeatedly hinted that investors could expect updates "soon" on the company's AI roadmap, but has not provided specific details about monetization strategies or timelines. This lack of clarity has tested investor patience, particularly given the company's history with the metaverse.

Meta ended the most recent quarter with $90.3 billion in cash and marketable securities versus $83.7 billion in debt. While this appears to provide a financial cushion, the company's growing off-balance-sheet liabilities make this comparison less reliable than it appears on the surface. Management did not increase the high end of their capital expenditure guidance but did increase the low end narrowly to $130 billion, signaling that heavy AI infrastructure spending will continue.

The fundamental question for investors is whether Meta's optionality in AI represents genuine upside or merely a distraction from its core business. If the company succeeds in monetizing AI through any of its multiple paths, or if it returns to generating strong free cash flow by moderating capital expenditure, a dramatic re-rating could occur. Until then, the market appears content to value Meta's advertising business cheaply, treating the company's AI ambitions as a speculative bet rather than a strategic advantage.