In response to widespread concerns regarding Meta's workforce reductions, cuts to Reality Labs funding, and postponement of its foundational model release, Goldman Sachs offers a different interpretation. The bank views these actions not as a defensive contraction under AI investment pressure, but as a proactive strategic reallocation of resources from less efficient existing operations toward high-value computing needs. This realignment is seen as creating potential for upward revisions to long-term earnings per share (EPS).

Recent reports from multiple media outlets suggest Meta's management is considering three major initiatives: significant layoffs and organizational restructuring, reduced spending within the Reality Labs division, and a delay in the public release of the latest foundational model from its Super Intelligence Lab. Following these reports, Meta's stock price has declined approximately 20% since the day after its Q4 earnings report on January 29, contrasting with a roughly 7% drop in the S&P 500 index over the same period.

In a research report dated March 22, a Goldman Sachs team led by analyst Eric Sheridan stated that these three potential measures do not substantially exceed the forward-looking framework provided by management during the recent earnings call. A more critical assessment from the bank is that any cost-saving initiatives should be viewed as Meta's effort to rebalance efficiency and growth across the entire company. Resources freed up by these actions are expected to be primarily redirected toward AI growth investments, particularly those centered on computing power demand. Goldman Sachs also indicated that if these measures are implemented, Meta could potentially return to its historical pattern of providing conservative initial annual guidance followed by consistent upward revisions throughout the year.

Goldman Sachs maintains a Buy rating on Meta with a 12-month price target of $835. The firm quantified the potential EPS impact of cost optimization through three scenario analyses, concluding that Meta's current cost structure possesses significant flexibility. Rebalancing efforts focused on efficiency and growth are seen as capable of driving sustained positive EPS revisions.

Regarding layoffs, reports from Reuters indicate Meta is planning significant workforce reductions. Goldman Sachs quantified this in Scenario 1: assuming a 15% year-over-year reduction in total headcount by the end of 2026, followed by 5% growth in both 2027 and 2028, alongside approximately 6% annual growth in per-employee non-depreciation and amortization expenses. Under these assumptions, compared to Goldman's baseline forecast, Meta's 2026 GAAP EBIT and EPS would see a boost of over 10%, with the uplift exceeding 20% for 2027-2028. Goldman emphasizes that the essence of this scenario is not simply reducing headcount, but reflects a broader industry trend observed since mid-December 2025: companies are increasingly combining workforce adjustments with a strategic shift toward concentrating resources on AI and machine learning computing roles. In other words, Meta's layoffs are seen as closer to "replacing personnel"—substituting existing roles with more technically skilled talent possessing AI backgrounds—rather than shrinking the overall workforce scale. Goldman thus characterizes this as a proactive talent restructuring initiative, not a reactive cost-cutting measure driven by pressure.

Concerning Reality Labs, Bloomberg reported that Meta plans adjustments to core products within the division, which some initially interpreted as a complete exit from the business segment. Goldman Sachs considers this interpretation inaccurate, citing subsequent official statements from Meta clarifying that the adjustments primarily target reducing investment in legacy VR products like Horizon Worlds. The company reportedly remains highly committed to the broader spatial computing vision, with no change in strategic direction. In Scenario 2, Goldman assumes annual Reality Labs expenses are reduced by a high-single-digit to low-double-digit percentage, with 10% of the savings reinvested into the business. The potential EPS uplift for 2026-2028 under this scenario falls within a low- to mid-single-digit percentage range, indicating relatively modest but positive flexibility. Goldman's assessment is that Meta's move essentially lowers the priority of traditional VR within Reality Labs to free up more resources for AI integration and Augmented Reality (AR) development.

Regarding the delay in the foundational model release, reported by The New York Times, Goldman Sachs states that this timeline has never exceeded its expectations. Given that the Super Intelligence Lab was established in mid-2025 and has been operational for less than a year, compared to established frontier model organizations like Google DeepMind, OpenAI, and Anthropic which have been developing for years, Meta likely requires at least 9 to 12 months to demonstrate initial results in the public domain. Goldman reaffirms its alignment with management's public guidance: the timeframe for the public release of foundational models and the deployment of key AI strategic pillars (such as computing infrastructure and AI agent products) is primarily focused on the second half of 2026 through 2027. The initial models are viewed as just the starting point, with iterative evolution expected over multiple years. The bank believes that comparing the progress pace of Meta's Super Intelligence Lab to market expectations for OpenAI or Anthropic, based solely on the time since establishment, involves a fundamental benchmark discrepancy.

Goldman Sachs constructed three "blue-sky scenarios" in its report to quantify the potential earnings impact of different cost optimization paths, explicitly stating these are not baseline forecasts and actual outcomes depend on multiple variables. Scenario 1 (adjusting headcount and per-employee costs) shows the most significant potential uplift, with EPS potentially increasing over 10% in 2026 and exceeding 20% in 2027-2028. Scenario 2 (reducing Reality Labs spending) offers more limited flexibility, providing a low- to mid-single-digit percentage EPS boost over the three years. Scenario 3 (zero growth in non-D&A operating expenses over the next three years) theoretically offers the greatest potential, potentially boosting 2026 EPS by over 30% and 2027-2028 EPS by approximately 40% to 50%. However, Goldman characterizes this as the least probable scenario, citing key constraints including persistent wage inflation for AI and machine learning roles and expected growth in total company headcount over a multi-year horizon.

Goldman Sachs concludes that Meta's existing cost structure provides substantial earnings flexibility. Any management actions aimed at balancing efficiency with growth investment are seen as having the capacity to drive ongoing upward revisions to EPS expectations in the coming years, rather than creating downward pressure.