Is Meta (META) Still a Buy in the AI Era? Reconciling Short-Term Valuation Concerns with Long-Term AI-Driven Growth

Meta Platforms (META) has long been a bellwether for the tech sector, but its recent performance in the AI era has sparked a critical debate: Is the stock still a buy, or has its valuation outpaced its fundamentals? With a forward P/E of 27.31 and a P/S of 8.78, Meta trades at a premium to Alphabet (P/E: 18.55, P/S: 6.14) but a discount to NVIDIA (P/E: 30x) and Microsoft (P/E: 30x). This valuation gap reflects diverging investor sentiment—optimism about Meta's AI-driven transformation versus skepticism about its ability to monetize these investments. Let's dissect the forces at play.

Earnings Outperformance and AI-Driven Momentum

Meta's Q2 2025 results were a masterclass in execution. Revenue surged 22% YoY to $47.52 billion, with EPS of $7.14 (up 38% YoY), far exceeding consensus estimates. The company's AI initiatives are already paying dividends: AI-powered ad tools like Advantage+ have boosted ad effectiveness by 5% on Reels, while 30% of advertisers now use AI-driven creative tools. These metrics underscore Meta's ability to leverage AI for incremental revenue, not just cost-cutting.

The company's AI infrastructure bets are equally aggressive. With $66–72 billion in 2025 capex, Meta is building “titan clusters” like Prometheus (1 GW in Ohio) and Hyperion (5 GW by 2030 in Louisiana). These facilities will power next-gen models like Llama 4, which is already available on Microsoft Azure and Amazon AWS. By open-sourcing its AI models, Meta is fostering a developer ecosystem that rivals the closed systems of Google and Microsoft. This strategy could unlock new revenue streams in enterprise AI and metaverse applications.

Valuation Metrics: A Double-Edged Sword

Meta's P/E and P/S ratios suggest a stock priced for growth, but they also highlight risks. At 27.31x forward earnings, Meta trades at a 20% premium to the S&P 500's 24.9x P/E. While this premium is justified by its AI roadmap, it leaves little room for error. For context, NVIDIA's 30x P/E reflects its dominance in AI chips and cloud infrastructure, whereas Meta's AI monetization is still nascent.

The company's PEG ratio of 1.72 is moderate, but it hinges on delivering 13.67% revenue growth in 2025. This depends on scaling AI-driven ad tools and monetizing its Reality Labs division, which reported $370 million in Q2 revenue (up 4.8% YoY). While promising, this pales against the $47.15 billion generated by its Family of Apps. Investors must ask: Can Meta replicate its ad success in AI hardware and enterprise services?

Regulatory Risks: The EU's Heavy Hand

Meta's AI ambitions face headwinds from regulatory overreach, particularly in the EU. The Digital Markets Act (DMA) has forced Meta to offer a “Less Personalized Ads” option, which could reduce ad effectiveness by 90% in Europe. This model, which relies on 10% less data for targeting, risks alienating advertisers and eroding Meta's 24.4% YoY European ad revenue growth.

The General Data Protection Regulation (GDPR) adds another layer of complexity. A €1.2 billion fine for unlawful data transfers to the U.S. (under Schrems II) remains a precedent, with the EDPB ordering Meta to suspend future transfers. While the EU-U.S. Data Privacy Framework (DPF) was introduced in 2023, legal challenges persist, creating uncertainty for transatlantic data flows. These risks could force Meta to repatriate data or invest in costly compliance measures, dampening margins.

Capital Intensity and Competitive Dynamics

Meta's AI strategy is capital-intensive. Its 2025 capex of $66–72 billion is nearly double 2024's $35–40 billion, with infrastructure costs expected to drive 2026 expenses. While this spending is critical for maintaining AI leadership, it pressures free cash flow (Q2: $8.55 billion) and operating margins (43% in Q2 vs. 37% in 2024). Competitors like Amazon (projected $111 billion in AI capex by 2025) and Microsoft ($30 billion) are also scaling aggressively, raising the bar for Meta's ROI.

However, Meta's open-source approach gives it an edge. By making Llama models freely available, it has attracted 1 billion monthly users and partnerships with Azure and AWS. This ecosystem-driven strategy could accelerate adoption and create network effects, unlike the closed ecosystems of Google and Microsoft.

Investment Thesis: Hold or Pivot?

Meta's AI investments are foundational, but its valuation and regulatory risks make it a cautious buy. The stock's 27.31x P/E is justified by its long-term potential but leaves little margin for error in the short term. Investors should consider the following:

1. Hold for AI Breakthroughs: If Meta delivers on Llama 4's enterprise adoption or monetizes smart glasses (via its 3% EssilorLuxottica stake), the stock could outperform.
2. Wait for a Pullback: A 15–20% correction in META's price could make the stock more attractive, especially if AI-driven ad revenue exceeds $3–6 billion by 2026.
3. Pivot to Undervalued AI Plays: For risk-averse investors, NVIDIA or Microsoft offer more mature AI monetization with lower regulatory exposure.

In conclusion, Meta remains a pivotal player in the AI arms race, but its valuation and regulatory challenges demand patience. The stock is best suited for investors with a 3–5 year horizon who can stomach near-term volatility. For those seeking immediate AI exposure, the landscape offers more undervalued opportunities.