Rippling unveiled AI Spend Console, a tool designed to help enterprises track and contain AI spendin...
The AMW Read
New enterprise AI governance tool signals a shift toward cost control in AI adoption, updating the finance/ops segment with a novel product response to a documented overspend problem.
Rippling unveiled AI Spend Console, a tool designed to help enterprises track and contain AI spending. The product maps individual employee, team, and role-level AI usage against productivity metrics, such as code review quality, to identify wasteful spending. Its launch follows Rippling's own internal crisis: in March, the company discovered it was on track to spend 40% of its R&D headcount budget on AI tokens, with month-over-month spend growing 80%. One engineer alone was spending $50,000 per month. The tool includes an AI gateway that routes prompts to the most cost-effective models, and dashboards that score spend against output.
The launch reflects a broader enterprise shift in 2026: after an initial wave of unrestricted AI adoption, companies are now demanding visibility and control over token costs. Rippling found that 10-15% of employees drove 60% of AI spend, and that defaulting to frontier models for all tasks was a major cost driver. By implementing caps, routing to cheaper models like Z.ai's GLM 5.2 (which Rippling benchmarked as nearly identical in performance to frontier models at 85% lower cost), and appointing "AI captains" to spread effective usage, Rippling cut its token spend from 40% to 15% of its R&D headcount budget, while maintaining near-peak usage volume. This validates the emerging category of AI spend management and gateway infrastructure, as enterprises move from experimentation to cost discipline.
For builders and investors, the implication is clear: AI cost governance is becoming a critical enterprise pain point. Startups offering usage analytics, model routing, or spend optimization for AI tools are well-positioned as enterprises scale internal AI adoption. For enterprises, the lesson is to negotiate hard caps with AI vendors early and to implement routing layers that match model choice to task complexity, rather than defaulting to the most expensive frontier options. The market for AI observability and cost control is likely to grow as token consumption continues to rise.