Meta’s Litigation Overhang Eases Following Settlement, While AI-Driven Advertising Growth and Computing Capacity Monetization Support the Risk-Reward Profile

Overview

The report focuses on two controversies weighing on Meta’s share price: youth social media litigation and the scale, pace, and returns of AI capital expenditure. Goldman Sachs believes the litigation settlement exchanges unknown risks for clearer operating rules, while AI-driven growth in the advertising business, new products, and monetization of surplus computing capacity provide upside optionality that is not yet fully reflected in the current valuation.

Core views

Meta’s recent share-price performance has lagged significantly: it is down 13% year to date, while the S&P 500 Index and Nasdaq 100 Index are up 12% and 16%, respectively. Goldman Sachs believes the primary pressure stems from two areas: the market continues to question the scale, pace, and returns of AI capital expenditure, while multiple lawsuits concerning Meta’s product design and its impact on younger users have raised concerns about potential liabilities, remedies, and even adjustments to the business model. Reports that plaintiffs were seeking US$1.4 trillion in penalties further amplified this non-fundamental tail risk. The latest settlement converts some unknown risks into defined obligations. Meta agreed to pay penalties of up to US$17.1 billion to 52 attorneys general representing US states and territories and the District of Columbia, and to modify how minors use the Family of Apps. Measures include a two-hour daily usage limit, nighttime access and school modes, productivity pauses, hidden like or reaction counts, beauty filters disabled by default, the option to select a non-personalized feed, and stronger parental supervision tools. Backend measures also include strengthening age verification and enforcement of age restrictions, collecting only enough data to verify the ages of users under 13, and enhancing age-appropriate content management. Product adjustments could still affect user growth, engagement, and monetization over the medium term, but Goldman Sachs believes the known penalties and clearer youth safety rules reduce previously difficult-to-quantify uncertainty. As litigation pressure recedes, Goldman Sachs believes investors can refocus on operating fundamentals and AI execution. The report divides Meta’s AI opportunities into new consumer and enterprise products, external monetization of computing capacity, and the application of AI to the core advertising business. Product opportunities for consumers, creators, and small and medium-sized businesses include paid AI subscriptions, enterprise and creator AI agents, internal and external AI infrastructure and software, and experimental consumer products from the AI incubator. Among these, “Business-in-a-Box” is viewed as a particularly attractive concept: Meta could leverage its existing commercial relationships with businesses and creators worldwide to provide small and medium-sized businesses with integrated automated services spanning advertising, marketing, customer support, content generation, and agentic commerce. The report believes these initiatives offer multiyear revenue growth potential, although some prospective contributions remain difficult to incorporate explicitly into revenue models. The core advertising business already provides more observable evidence of returns on AI investment. Goldman Sachs expects Meta’s advertising revenue to grow 27% in 2026 and 21% in 2027; these forecasts are approximately 18% and 28% higher, respectively, than its January 2025 estimates. This growth is significantly above Goldman Sachs’ forecast of an approximately 9% two-year compound growth rate through 2027 for global digital advertising expenditure excluding China. The report identifies user growth, improved engagement, advertising automation, and generative AI advertising creative as the primary transmission channels, and views continued upward revisions to advertising revenue forecasts as evidence that returns on AI capital investment have begun to emerge. External sales of computing capacity represent another option not yet fully priced in. Meta continues to prioritize computing capacity for its own AI objectives, but management mentioned the possibility of licensing surplus capacity externally during the second-quarter earnings call. Goldman Sachs expects the industry-wide computing capacity shortage could persist beyond 2028. Consequently, leasing idle capacity in the near term could generate revenue and strengthen market confidence in returns on current capital expenditure. Potential models include a “neocloud” model, under which bare-metal training or inference capacity is leased directly to foundation-model companies, and a “hyperscale cloud” model, under which a complete suite of distributed AI services is built for enterprise customers. The former could generate high-margin revenue with strong cash-flow conversion in the near term, but its sustainability, moat, and long-term returns on capital remain debatable. The latter offers a larger and more durable revenue opportunity but would require building enterprise sales and go-to-market capabilities while confronting more direct competition and an unclear competitive advantage. The computing capacity monetization scenario analysis is based on three variables: total capacity, the proportion allocated externally, and revenue per gigawatt. Goldman Sachs estimates that Meta’s total computing capacity will increase from 3.5 gigawatts at the end of 2025 to approximately 7 gigawatts at the end of 2026 and approximately 12.5 gigawatts at the end of 2027. On this basis, it estimates that external sales of surplus computing capacity could generate approximately US$9.3 billion to US$56.2 billion in revenue in 2027. Not all of this range represents incremental revenue because external allocation could displace internal use; the capacity that can be completed over the next 12 to 18 months, management’s willingness to enter such transactions, and transaction timing also remain uncertain. Goldman Sachs therefore frames this as upside optionality rather than certain revenue in its base forecast. The financial forecasts reflect the coexistence of rapid growth and large-scale investment. Goldman Sachs expects total revenue to rise from US$200.965 billion in 2025 to US$255.409 billion in 2026, US$310.507 billion in 2027, and US$370.837 billion in 2028, corresponding to growth of 27.1%, 21.6%, and 19.4%; earnings per share are expected to be US$31.65, US$31.39, and US$37.83, respectively. Over the same period, research and development expenses are expected to increase from US$57.372 billion to US$82.491 billion, US$98.989 billion, and US$106.908 billion, while the EBIT margin is projected to decline from 41.4% in 2025 to 34.0% in 2026 and 29.0% in 2027 before recovering to 32.0% in 2028, indicating that AI and infrastructure investment will weigh on margins over the medium term. Forecast price-to-earnings ratios are 18.2x for 2026, 18.4x for 2027, and 15.2x for 2028. Goldman Sachs believes the greatest determinant of subsequent share-price performance will be whether product launches, AI revenue, and advertising improvements can demonstrate sufficiently quickly that future computing capacity investment can generate attractive returns. If management continues to demonstrate monetization progress and improves visibility into the long-term economics of AI, Meta’s current major valuation discount could narrow significantly. The cadence of foundation-model releases over the coming months, particularly the Meta Connect event on September 23, is identified as a key catalyst in the report. Goldman Sachs reiterates its Buy rating and 12-month price target of US$725. The target price is derived by equally weighting two methods: first, applying a 26.0x EV/GAAP EBIT multiple to the forecast for the next 12 months plus one year; and second, using a modified discounted cash flow method that applies a 45.0x EV/free cash flow after stock-based compensation multiple to the forecast for the next 12 months plus four years and discounts it back three years. Relative to the current price of US$576.14, the target price implies 25.8% upside.

Analysis framework

The report first identifies the two primary pressures behind recent relative share-price underperformance and then separately assesses the impact of the litigation settlement on tail risk and business operations. It subsequently analyzes revenue opportunities across three paths—consumer and enterprise AI products, external monetization of computing capacity, and the application of AI to advertising—and constructs a 2027 scenario analysis using capacity, the proportion allocated externally, and revenue per unit of computing capacity. Finally, it derives the target price by combining financial forecasts, visibility into returns on capital expenditure, and two valuation methods.

Methodology notes

• Valuation MethodDCF Discounted Cash Flow

  Modified Discounted Cash Flow Valuation

  The report applies a 45.0x EV/free cash flow after stock-based compensation multiple to the forecast for the next 12 months plus four years and discounts it back three years; the result is equally weighted with the relative valuation result to derive the target price.

• Valuation Method

  EV/GAAP EBIT Relative Valuation

  The report applies a 26.0x enterprise value-to-GAAP EBIT multiple to the forecast for the next 12 months plus one year to reflect the market’s valuation of Meta’s operating profit.

• Industry/Sector Analysis FrameworkSupply-demand framework

  AI Computing Capacity Supply-Demand Gap Analysis

  Based on constrained industry computing capacity supply, robust external demand, and the possibility that tight conditions could persist beyond 2028, the report analyzes the commercial opportunity for Meta to lease surplus computing capacity.

• Industry/Sector Analysis Framework

  External Computing Capacity Monetization Scenario Analysis

  The report simultaneously adjusts total computing capacity, the proportion allocated externally, and revenue per gigawatt, while referencing public comparable data from neocloud and hyperscale cloud companies and similar licensing transactions to estimate the 2027 revenue range.

Asset mapping & comparison

Structured mapping from thesis to named assets (strengths, weaknesses, peers, risks).

• Meta Platforms Inc. (META.US)

  The litigation settlement reduces significant non-fundamental uncertainty, while AI-driven advertising improvements and the monetization of surplus computing capacity provide potential upside optionality.

  Strengths

  Meta has commercial relationships spanning businesses and creators worldwide, a mature Family of Apps platform, and foundations for monetizing advertising automation, generative creative tools, and AI products.

  Weaknesses

  Large-scale AI and infrastructure investment is expected to weigh on medium-term margins, while some new products and external computing capacity revenue have yet to generate contributions that can be modeled reliably.

  Comparison

  Goldman Sachs expects Meta’s advertising revenue to grow 27% and 21% in 2026 and 2027, respectively, significantly above the approximately 9% two-year compound growth rate through 2027 for global digital advertising expenditure excluding China.

  Risks

  Risks include competition for users and advertising, prolonged pressure on margins from long-term investment, regulatory and antitrust pressure, and the failure to monetize new-business optionality.

Key data

• 12-Month Price TargetUS$725.00Price target reiterated, implying 25.8% upside relative to the current price
• Current PriceUS$576.14Price basis used in the report
• Year-to-Date Share-Price Performance-13%Over the same period, the S&P 500 Index rose 12% and the Nasdaq 100 Index rose 16%
• Settlement PenaltiesUp to US$17.1 billionPayable to 52 attorneys general representing US states and territories and the District of Columbia
• Penalties Reportedly Sought by PlaintiffsUS$1.4 trillionPreviously heightened market concerns about litigation tail risk
• Advertising Revenue Growth Forecast2026 +27%, 2027 +21%Approximately 18% and 28% above the January 2025 forecasts, respectively
• Growth in Global Digital Advertising Expenditure Excluding ChinaApproximately 9% two-year compound growth rate through 2027Below Goldman Sachs’ growth forecast for Meta’s advertising revenue
• Total Computing Capacity3.5 gigawatts at the end of 2025, approximately 7 gigawatts at the end of 2026, and approximately 12.5 gigawatts at the end of 2027Used in the external computing capacity monetization scenario analysis
• Potential External Computing Capacity Revenue in 2027Approximately US$9.3 billion to US$56.2 billionDepends on externally allocated capacity and revenue per gigawatt and may not be entirely incremental revenue
• Total Revenue ForecastUS$255.409 billion in 2026, US$310.507 billion in 2027, and US$370.837 billion in 2028Corresponding to year-over-year growth of 27.1%, 21.6%, and 19.4%
• Earnings per Share ForecastUS$31.65 in 2026, US$31.39 in 2027, and US$37.83 in 2028Expected to decline 0.8% year over year in 2027 and grow 20.5% in 2028
• Forecast Price-to-Earnings Ratio18.2x in 2026, 18.4x in 2027, and 15.2x in 2028Calculated using the price basis adopted in the report

Impact & implications

The report believes the settlement does not eliminate the potential medium-term impact of product adjustments on engagement and monetization, but it converts difficult-to-quantify litigation risks into defined costs and operating requirements. Market attention will next shift to whether AI capital expenditure can generate verifiable returns through advertising growth, new product revenue, and external computing capacity monetization; if visibility into returns improves, Meta’s valuation discount could narrow.

Risks

• Product adjustments for minors could affect user growth, engagement, or monetization over the medium term.
• Existing and emerging competitors across the internet, media, and commerce sectors could compete for users, time spent, and advertising budgets.
• Large investments in long-term projects could keep operating margins below forecasts for longer than expected.
• Incremental capital returns to shareholders could be low or absent in the future.
• Regulatory scrutiny and changes in industry rules could weaken the prospects of the existing business model.
• Antitrust scrutiny could hinder the acquisition of talent and product innovation through mergers and acquisitions or lead to a breakup and impairment of synergies from previous acquisitions.
• Upside optionality in information services, social commerce, Oculus, and the metaverse may fail to achieve commercialization.
• The global macroeconomic environment and investor risk appetite for growth stocks could cause share-price volatility.

What to watch

• Monitor progress in foundation models, AI products, and the long-term AI strategy disclosed at the Meta Connect event on September 23.
• Monitor the cadence of foundation-model results and new product launches over the coming months.
• Monitor whether advertising improvements, new AI revenue, and product monetization can demonstrate attractive returns on computing capacity investment.
• Monitor whether management licenses surplus computing capacity externally and whether it adopts a neocloud model or a hyperscale cloud model.
• Monitor changes in growth, engagement, and monetization metrics among underage users following implementation of the settlement requirements.