Why do companies face rising AI costs?

Costs surge due to uncontrolled use of expensive models for all tasks, lack of cost transparency, and absence of optimization tools. Many employees apply advanced models even for simple operations, increasing expenses.

How does AI Spend Console reduce costs?

The tool analyzes expenses by employees and teams, correlates them with productivity, and automatically selects the most cost-effective AI models for specific tasks. This reduces costs without limiting AI usage.

Which AI models does Rippling use for optimization?

Rippling employs multiple models, including Z.ai's GLM 5.2, which is 85% cheaper than premium alternatives but delivers comparable performance. It also uses a proprietary gateway system for request routing.

← All news

Artificial intelligence

Rippling Unveils AI Spend Console to Control Enterprise AI Costs

August 7, 2026

Photo: TechCrunch

Quick answer

HR platform Rippling has introduced AI Spend Console to control and optimize corporate AI spending.

HR platform Rippling has introduced AI Spend Console, a tool designed to monitor and optimize corporate AI spending. The solution emerged after the company faced a sharp rise in token costs: in March 2024, AI expenses reached 40% of the R&D department's budget, equivalent to the salaries of 40% of its employees.

Analysis revealed that 10–15% of employees generated 60% of all AI costs, with one engineer spending $50,000 monthly. The core issue was a lack of control—employees defaulted to using the most expensive models for all tasks, even simple ones. Instead of banning AI, Rippling implemented strict limits and negotiated fixed-rate contracts with providers like OpenAI and Anthropic.

The new AI Spend Console integrates Rippling’s proprietary gateway system, which automatically selects the most cost-effective model for each task. For routine operations, it uses China’s GLM 5.2 model, which is 85% cheaper than premium alternatives but matches their performance. As a result, the company reduced token costs by 62% while maintaining AI usage levels.

The tool also provides dashboards that display expenses by employees, teams, and roles, correlating them with productivity metrics. This helps identify inefficient AI usage and optimize workflows. For example, Rippling found that some high-spending engineers frequently received code revision requests, indicating low work efficiency.

Looking ahead, Rippling plans to expand the tool’s use across other departments, including customer-facing teams. However, AI access may become restricted for employees whose productivity cannot be measured. This could mean corporate AI access will no longer be universal, like Slack or email, and will instead be granted only to those who demonstrate measurable efficiency.

Common questions

Why do companies face rising AI costs?: Costs surge due to uncontrolled use of expensive models for all tasks, lack of cost transparency, and absence of optimization tools. Many employees apply advanced models even for simple operations, increasing expenses.
How does AI Spend Console reduce costs?: The tool analyzes expenses by employees and teams, correlates them with productivity, and automatically selects the most cost-effective AI models for specific tasks. This reduces costs without limiting AI usage.
Which AI models does Rippling use for optimization?: Rippling employs multiple models, including Z.ai's GLM 5.2, which is 85% cheaper than premium alternatives but delivers comparable performance. It also uses a proprietary gateway system for request routing.

Dzen feed: /feed/dzen.xml · RSS: /feed.xml