By Hemang Davé and Reinier Aerdts

FinOps is approaching a turning point in its maturity journey.

Nearly three-quarters (75%) of the Forbes’ Global 2000 companies now use the financial management methodology to maximize the business impact of their cloud and hybrid cloud technology.¹ Yet with global cloud-related spending growing 26% annually,² many FinOps teams are seeking avenues for more value.

AI and generative AI offer such a way forward.

Applying these advanced technologies to an established FinOps framework can amplify the speed, accuracy and effectiveness of traditional practices. These enhancements should, in turn, put your FinOps program on a path for long-term, sustainable growth.

Potential use cases for AI in FinOps

As a recognized strategy, integrating AI and generative AI with FinOps is still in its infancy. However, we expect adoption rates to increase as more organizations explore the approach. Use cases may include:

• Automated cost anomaly detection
  With traditional FinOps, anomaly detection often relies on manual reviews or static threshold-based alerts. Though effective, this method can be time-consuming, less scalable, and prone to missing subtle or emerging patterns in spending behavior.
  
  On the other hand, AI and generative AI tools integrated with FinOps can continuously monitor spending patterns and automatically flag unusual activities. This automated approach enables more comprehensive analysis of large datasets, which speeds detection and helps reduce cost spikes or anomalies in real time.
  
  
• Enhanced forecasting and budgeting
  Standard FinOps methodologies depend on historical averages and manual estimations for forecasting and budgeting. However, this approach can lack the agility and precision needed to adapt to dynamic cloud usage patterns and external market influences like economic shifts and changes in consumer preferences.
  
  AI- and generative AI-enabled FinOps can analyze historical data, usage trends and external factors to provide more accurate predictions for forecasting and budgeting.³ FinOps teams can also use AI tools to simulate different scenarios to help anticipate various outcomes, improving financial management and resource planning.
  
  
• Dynamic resource optimization
  Traditional FinOps uses scheduled reviews and manual rightsizing to optimize resources. Without automation, FinOps teams can sometimes delay making adjustments and miss opportunities for real-time efficiency gains.
  
  When FinOps teams deploy AI and generative AI, the tools dynamically adjust cloud resource allocations based on real-time demand and workload requirements. This process, which may include shutting down unused or underutilized resources, helps maximize cloud resource usage and costs.
  
  
  

Benefits of applying AI to FinOps

Regardless of the application or industry, integrating AI and generative AI tools with FinOps principles should enable organizations to increase productivity, improve decision-making and lower costs more dramatically than with FinOps alone.⁴ These results are driven by:

• Automation of routine tasks like invoicing, auditing and account reconciliation
• Real-time reporting and financial insights for faster adjustments and decision-making
• Predictive analytics to forecast trends, predict cash flow and identify potential risks
• Compliance and governance through automated enforcement and detailed reporting

To appreciate the potential of AI- or generative AI-enabled FinOps, imagine a hypothetical video-on-demand platform with millions of global subscribers and a vast content library.

The platform experiences unpredictable spikes in cloud usage during major content releases and live events. Although the platform’s parent company uses FinOps to help control cloud costs, traditional practices struggle to keep up with real-time demand. These limitations result in overprovisioned resources during low-traffic periods, cost spikes during high-demand events, delayed anomaly detection and budget overruns.

If the parent company launched an AI-enhanced FinOps solution, it would allow real-time resource optimization, continuous anomaly detection and dynamic cost control. Teams might also enable automated natural language summary reporting to generate actionable insights based on key trend data. Collectively, these capabilities should deliver a higher FinOps return on investment (ROI) through automation, predictive insights and cloud cost optimization, ultimately creating better experiences for the company’s customers.

How to prepare for AI and FinOps implementation

Deploying AI and generative AI within a FinOps framework is a collaborative endeavor that requires finance, engineering and business teams to work outside their usual scope. These steps can facilitate the process:

1. Prepare financial data AI and generative AI insights are only as good as the information used to train the models, so establish governance protocols to manage the quality, use, security and compliance of organizational data. Once these guardrails are in place, collate data from various sources (business, IT, partner ecosystem) into a data lake where it can be easily accessed and analyzed. FinOps teams can then identify patterns and trends in the data, allowing them to make more informed decisions and improve outcomes.
   
   
2. Select robust tools
   When choosing AI-equipped tools for your FinOps program, look for solutions that offer predictive analytics to forecast costs and usage, automation to save time and reduce errors, and natural language processing (NLP) so non-technical users can query data. Applications should also integrate with existing FinOps platforms and scale with your organization’s needs while adhering to data standards and regulatory requirements.
   
   
3. Develop AI and generative AI models
   After preparing your data and implementing software, FinOps teams can create and train outcome-based AI and generative AI models tailored to specific business needs. This process includes defining problems and goals, selecting and training the appropriate algorithms, evaluating performance, and refining models to achieve desired outcomes.
   
   
4. Train cross-functional teams
   Provide cross-functional training in FinOps and AI to help stakeholders understand each other’s roles and how they contribute to common goals and collective success. Education can include formal training, mentoring, coaching, self-directed learning and guild participation. This multi-disciplinary approach accelerates implementation and helps ensure technical and financial insights drive all decision-making.
   
   
5. Monitor, measure, adjust and iterate
   To determine the business impact of integrating AI and generative AI into FinOps, set up protocols to routinely assess cost savings, productivity increases, risk reduction and other key performance indicators. You can then use this data to refine processes and enhance FinOps performance as AI and generative AI evolve.

Choosing your way forward

The road to FinOps maturity is paved with opportunities. Your FinOps team can take the traditional route and realize significant cloud-related savings. Or, you can blaze a new trail by combining AI and generative AI with established FinOps methodologies. This direction is still being plotted, but it should eventually lead to even greater returns on your FinOps investment.

Hemang Davé and Reinier Aerdts are Client Technical Leaders for the Technology, Media and Telecommunications market at Kyndryl, a member of the FinOps Foundation and FinOps Foundation Governing Board.

1 FinOps breaks out of the cloud, CIO, March 2025
2 Shared cloud infrastructure spending continues to accelerate, fueled by AI-related spending, IDC, June 2024
3 Key FinOps trends for CFOs to watch in 2025, CFO Dive, January 2025
4 Companies using AI-enabled tools for FinOps save more than 20%, BusinessWire, October 2023