1. MuSpark and Meta Super Intelligence Labs launch. Meta launched its Muse family of AI models this quarter, with MuSpark as the first release from the newly formed Meta Super Intelligence Labs (MSL). MuSpark now powers Meta AI across all Family of Apps surfaces — Facebook, Instagram, Messenger, WhatsApp, and the standalone Meta AI app. Prior to this quarter, Meta AI ran on older internal models and third-party APIs. Li reported double-digit percentage increases in Meta AI sessions per user following the broad MuSpark rollout.¹ Zuckerberg stated the company is already training more advanced successor models — establishing MuSpark as the first step on what he called a scaling ladder rather than a destination.

"Over the past 10 months, we have built the strongest research team in the industry and established the scientific and technical foundations to scale very advanced models. Spark is just one step on that scaling ladder, and we are already training even more advanced models."
— Mark Zuckerberg, CEO, Q1 FY2026 Earnings Call¹

2. CapEx raised a second consecutive quarter; contractual commitments up $107 billion. The revised 2026 CapEx guidance of $125–$145 billion is $10 billion higher at the midpoint than the Q4 2025 guidance of $115–$135 billion, which was itself a major raise from original 2026 planning levels.¹ Management attributed the increase primarily to higher memory component pricing, with a secondary contribution from data center costs tied to future-year capacity. Separately, multiyear cloud agreements and infrastructure purchase contracts drove a $107 billion increase in contractual commitments in a single quarter — a figure that compounds the investor concern about the capital intensity trajectory into 2027 and beyond.⁶

3. Planned headcount reduction announced for May 2026. Li disclosed that Meta informed employees internally of a planned reduction to its workforce in May. This follows a 1% sequential headcount decline already visible in Q1 (77,900 employees vs. Q4). The framing — a leaner operating model to move faster and offset infrastructure investment — is consistent with Zuckerberg's remarks about AI enabling small teams to ship products that previously required much larger groups. The workforce action represents a structural reallocation of capital from labor to compute rather than a response to revenue weakness.¹

4. AI monetization metrics crossed quantifiable thresholds. Two monetization datapoints are new this quarter: the value optimization suite crossed a $20 billion annual revenue run rate (more than doubling year-over-year), and partnership ads crossed a $10 billion annual run rate (also doubling YoY).¹ Business AI conversations on WhatsApp and Messenger reached 10 million per week, up 10x from 1 million at the start of 2026. These represent concrete adoption data points, though neither the value optimization run rate nor the partnership ads run rate is broken out from the overall advertising revenue figure for independent verification.

Meta Platforms Q1 FY2026: The CapEx Shock That Overshadowed a 33% Revenue Beat

The Power Delivery Gap