By Maksym Misichenko · Yahoo Finance ·
What AI agents think about this news
The panel is divided on the significance of the Anthropic deal for Meta's future. While some see it as a potential revenue stream and validation of Meta's AI capabilities, others caution about energy risks, regulatory hurdles, and the 'coopetition' trap.
Risk: The 'coopetition' trap and potential commoditization of Meta's software advantage.
Opportunity: Potential high-margin recurring revenue stream from Meta Compute.
This analysis is generated by the StockScreener pipeline — four leading LLMs (Claude, GPT, Gemini, Grok) receive identical prompts with built-in anti-hallucination guards. Read methodology →
Meta Platforms (META) has spent years pouring tens of billions of dollars into artificial intelligence (AI). Investors have largely accepted those investments because advertising profits have remained exceptionally strong, but one question has lingered over the stock: When will Meta actually make money from its AI infrastructure?
That question is becoming much more important after reports surfaced that Anthropic is discussing a compute-leasing agreement with Meta that could be worth as much as $10 billion over two years. If completed, the arrangement would represent one of the first major examples of Meta generating direct revenue from the massive AI data centers it has been building.
Meta will report its second-quarter earnings on July 29. As always, investors will likely be looking for signs that the company's enormous AI investments are beginning to create entirely new revenue streams rather than simply supporting its advertising business.
Meta Platforms Stock Still Looks Reasonably Valued
Despite the rebound in July, Meta Platforms' valuation remains relatively reasonable compared with many AI leaders.
META stock currently trades at roughly 21 times trailing earnings and about 21.4 times forward earnings, below its five-year average multiple of around 26 times. Meta also carries a PEG ratio near 1.07 times, suggesting expected earnings growth largely offsets its current valuation.
After falling roughly 15% during the first half of the year, Meta has recovered sharply. Shares are now down 8% year-to-date (YTD) after rallying this month, with investors embracing the company's evolving AI strategy.
If Meta succeeds in generating recurring revenue from AI infrastructure, today's valuation could prove more attractive than it initially appears.
Meta's AI Investments Are Starting to Change the Narrative
Investors have spent much of the year questioning Meta's aggressive spending plans.
Management expects to spend between $125 billion and $145 billion in capital expenditures during 2026, almost double the roughly $72 billion invested last year. That level of spending initially pressured META stock as investors wondered whether returns would justify the enormous investment.
The picture has started to improve. Bloomberg reported earlier this month that Meta is developing Meta Compute, a cloud computing platform that would rent AI infrastructure to outside customers. The report immediately changed investor sentiment because it suggested Meta could eventually compete with the likes of Amazon's (AMZN) AWS, Microsoft (MSFT) Azure and Alphabet's (GOOGL) Google Cloud rather than remaining solely an advertising company.
News that Anthropic may lease computing capacity from Meta reinforces that strategy. Instead of viewing AI infrastructure purely as a cost, investors are beginning to see it as a potential long-term revenue generator.
Meta's Business Continues Delivering Strong Results
While attention has shifted toward AI monetization, Meta's core business continues to produce impressive financial results.
During the first quarter, Meta reported EPS of $10.57, comfortably ahead of analysts' expectations of $6.65. Revenue climbed 33% year-over-year (YOY) to $56.3 billion, also exceeding Wall Street forecasts. The company has now beaten consensus earnings estimates in each of the past four quarters.
Wall Street expects another solid report on July 29, forecasting revenue of approximately $60.2 billion and earnings of $7.13 per share. Bank of America is even more optimistic, projecting $60.6 billion in revenue and EPS of $7.50, driven by healthy advertising demand and AI-powered improvements across platforms.
Beyond earnings, investors will likely focus on updates regarding Meta Compute, AI infrastructure monetization, and any changes to management's capital spending outlook.
Wall Street Remains Bullish on Meta Platforms Stock
Analysts remain optimistic about Meta's long-term outlook. Based on 54 analysts with coverage, META stock carries a consensus "Strong Buy" rating. The average price target of $824.29 suggests potential upside of about 36% from current levels.
Meta was recently called one of the strongest beneficiaries of AI spending by Goldman Sachs. The firm has a "Buy" rating on META stock with an $830 price target. Analysts at Morgan Stanley also have an "Overweight" rating with a target price of $775, believing that worries about AI investing returns have already been priced into shares.
Bank of America is still among the largest bulls with an $835 price target, keeping a "Buy" rating for the stock. The firm suggests that investors still don't appreciate the long-term worth of Meta's growing AI systems, and that compute leasing and other prospects might come and slowly change the valuation of the firm.
The proposed Anthropic agreement is still in talks and may not be realized if it misses approval in the near future. However, if Meta eventually transforms its AI infrastructure into a valuable cloud company, investors will look at it differently. Meta Platforms could eventually be seen as one of the world's biggest AI infrastructure companies rather than just a social media giant.
On the date of publication, Nauman Khan did not have (either directly or indirectly) positions in any of the securities mentioned in this article. All information and data in this article is solely for informational purposes. This article was originally published on Barchart.com
Four leading AI models discuss this article
"The Anthropic deal marks the inflection where Meta’s AI infrastructure shifts from cost center to verifiable revenue stream, supporting a valuation re-rating."
The $10B Anthropic compute-leasing deal, if closed, is the first tangible proof that Meta can monetize its $100B+ AI capex beyond its core ad business. At 21.4x forward earnings with 19-20% expected EPS growth and a PEG ~1.07, the stock is not expensive relative to hyperscalers. Q2 earnings on July 29 and any Meta Compute updates could catalyze a re-rating toward 26x multiple, implying 20-25% upside from current levels. However, $125-145B 2026 capex guidance is still enormous and front-loaded; actual monetization timing and margins remain uncertain.
The deal is still only in discussion, may never close, and even if realized represents <5% of Meta’s current revenue run-rate while adding massive incremental energy, regulatory, and execution risks; the article glosses over how many years of negative free-cash-flow drag could occur before AI infrastructure becomes accretive at scale.
"The potential Anthropic deal signals a structural re-rating of Meta from a social media advertising firm to a foundational AI cloud infrastructure provider."
The market is mispricing Meta’s pivot from a pure-play ad platform to a diversified compute provider. A $10 billion deal with Anthropic isn't just revenue; it validates Meta’s proprietary infrastructure stack (Llama/PyTorch) as a viable alternative to AWS or Azure. With a PEG ratio of 1.07, the stock is priced for a deceleration that isn't showing up in the ad-tech stack. If Meta successfully monetizes its H100/B200 clusters via 'Meta Compute,' they effectively turn a massive CapEx burden into a high-margin recurring revenue stream, justifying a P/E expansion toward 25x-28x as the market re-rates them as a cloud-infrastructure utility.
Meta lacks the enterprise-grade support, security, and global data center footprint of Microsoft or Amazon, making them a 'compute vendor of last resort' rather than a true cloud competitor.
"Meta's advertising profits are strong enough to justify the stock today, but the $125-145B capex bet on infrastructure monetization remains unproven and could destroy shareholder value if adoption stalls or competition intensifies."
The Anthropic deal is real optionality, but the article conflates two separate narratives: (1) Meta's advertising business is genuinely strong—Q1 EPS beat by 59%, and that's the actual earnings engine today; (2) Meta Compute is speculative infrastructure monetization years away. The $10B Anthropic agreement, if real, is ~1.4% of Meta's projected 2026 capex and doesn't meaningfully change near-term returns on the $125-145B spend. The valuation at 21x forward isn't cheap relative to 19-21% expected growth; it's fair. The real risk: if AI infrastructure becomes commoditized (AWS, Azure, Google Cloud already have scale and customer lock-in), Meta's capex becomes stranded cost, not a moat.
If Meta Compute actually scales to rival AWS/Azure, the infrastructure business could be worth $50-100B in NPV by 2030—justifying today's capex and re-rating the stock 40-60% higher. The article may be underweighting this tail scenario.
"Meta's upside from AI compute monetization depends on durable cloud margins and scalable enterprise traction, which remain unproven."
The Anthropic lease hints at a potential revenue stream from Meta Compute, but the path is far from certain. Cloud markets are dominated by AWS, Azure, and Google Cloud, with tight margins and relentless price competition; Meta would need more than capacity—it's about winning enterprise trust, security, ecosystems, and a scalable go-to-market. A $10B two-year rent could uplift top-line only if adoption scales meaningfully, yet the ROI must clear a high hurdle given Meta’s massive capex plan for 2026 ($125-145B). Sentiment may already reflect AI upside, but the actual profits hinge on durable cloud margins and a sustainable enterprise business, not just a narrative shift.
The bear case is that even with a successful compute lease, the cloud business may yield slim margins for years amid fierce competition and heavy capex, and Anthropic deal may not materialize at all or could be smaller than implied.
"Power availability and regulatory bottlenecks, not customer revenue, are the real near-term gating factor for Meta Compute."
Claude correctly flags the 1.4% scale but understates energy risk. Hyperscale AI clusters already consume gigawatts; adding 2026 capex at $125-145B implies another 5-7 GW of incremental demand. Power contracts, not just revenue deals, become the binding constraint. Nobody has modeled the regulatory or permitting delays this creates across multiple states and Europe.
"Leasing compute to Anthropic creates a strategic paradox where Meta subsidizes the infrastructure for its own direct AI software competitor."
Grok is right on energy, but everyone is missing the 'coopetition' trap. If Meta leases compute to Anthropic, they are effectively subsidizing their own primary competitor in the LLM space. This isn't just a cloud utility play; it's a strategic paradox. If Llama remains open-source while Meta builds the hardware moat for Anthropic, they risk commoditizing their own software advantage. The capital isn't just stranded; it's potentially financing the erosion of their own AI leadership.
"Meta is potentially trading high-margin software optionality for low-margin hardware commoditization under the guise of infrastructure monetization."
Gemini's coopetition trap is the sharpest observation here, but it cuts both ways. Meta leasing to Anthropic could be defensive—starving OpenAI/Microsoft of capacity while keeping Llama's ecosystem alive. However, Gemini assumes Llama remains differentiated; if open-source models commoditize, Meta's hardware moat becomes their only edge. The real risk: they're betting infrastructure margins beat software margins, which historically they don't in cloud.
"Regulatory and grid-connection delays could push Meta Compute's ROI horizon far beyond 5–7 years, making near-term upside highly contingent on deployment milestones and energy-contract certainty."
Responding to Grok: energy is necessary but not sufficient — the bigger risk is regulatory and grid-connection delays that could push the ROI horizon on the 125-145B capex well beyond 5–7 years. Multi-jurisdiction permitting, PPA tightening, and price swings in power markets could erode margins regardless of a potential Anthropic lease. Even with monetization, the time-to-value remains long, making a near-term re-rating contingent on credible deployment milestones and energy-contract certainty.
The panel is divided on the significance of the Anthropic deal for Meta's future. While some see it as a potential revenue stream and validation of Meta's AI capabilities, others caution about energy risks, regulatory hurdles, and the 'coopetition' trap.
Potential high-margin recurring revenue stream from Meta Compute.
The 'coopetition' trap and potential commoditization of Meta's software advantage.