A practitioner-oriented post by a SaaS CFO, drawing on ICONIQ Capital's 2026 State of AI survey of roughly 300 executives, quantifies the gross-margin gap between traditional SaaS (70–80%) and AI-heavy SaaS products (~52% expected in 2026), attributing the compression to variable inference, model-routing, and vector-database costs that scale with usage rather than with seats. The piece also cites Salesforce Q4 FY2026 earnings data showing Agentforce ARR at $800M, up 169% YoY, with nearly 20 trillion tokens processed, illustrating that AI COGS is now material at enterprise scale. For financial institutions, this structural margin compression, if it persists across the app layer, bears directly on revenue quality, valuation multiples, and the durability of SaaS business models as AI usage grows.
291 other tracked developments touching the same claims.