Rippling unveils AI Spend Console to curb runaway corporate token costs

HR software provider Rippling has launched the AI Spend Console, a new governance tool designed to monitor and control corporate expenditure on artificial intelligence. The product was developed after the company identified that its own AI token usage was escalating by 80 per cent month-on-month, with costs consuming up to 40 per cent of its research and development headcount budget. By deploying an AI gateway to direct prompts to more cost-effective models and establishing spending caps with providers such as OpenAI, Anthropic, and Cursor, Rippling reduced its token spend to approximately 15 per cent of its R&D budget while maintaining high usage volumes.

The initiative began in March when Chief Product Officer Matt MacInnis and CFO Adam Swiecicki reviewed internal data during an executive meeting. The analysis revealed that spending was on track to equal 40 per cent of the compensation paid to the R&D unit. Further investigation showed that 10 to 15 per cent of employees drove 60 per cent of total AI spend, with one engineer alone accounting for $50,000 in monthly costs. In July, the company processed 600 billion AI tokens, highlighting the scale of the issue before intervention.

Rippling implemented an AI gateway to route prompts to the most cost-effective models for specific tasks, rather than defaulting to expensive frontier options. The company benchmarked various models and found that SpaceX’s Grok was the all-around leader, while Z.ai’s GLM 5.2 was 85 per cent cheaper with nearly identical performance for coding tasks. This strategy allowed the company to maintain high token volumes while significantly lowering costs.

The AI Spend Console links token consumption to productivity metrics, such as lines of code and pull requests, to determine if spending correlates with genuine output. Rippling appointed “AI captains” to assist employees in using AI effectively, initially focusing on engineering but expanding to customer onboarding teams. The tool includes dashboards that score attributes such as prompts per day combined with work output, helping managers identify high-spend individuals who may not be delivering proportional value.

As a result of these measures, Rippling dropped its token spend from 40 per cent of its headcount budget to about 15 per cent without curtailing overall usage. The cost of token spend in July was 37 per cent of the cost in April, despite similar token volumes. The AI Spend Console is included for Rippling’s HR subscribers, with additional usage-based costs, and can be purchased as a standalone product.