Agentic AI — artificial intelligence that can take autonomous action on tasks and decisions — is set to reshape operations for insurance, financial services, and more. From streamlining claims handling to automating underwriting tasks, or disputes, the potential to save time, reduce costs, and free up valuable capacity is enormous. According to McKinsey & Company’s report, The Future of AI in the Insurance Industry, “…leading insurers don’t see AI as just another efficiency tool—they recognize it as a fundamental driver of transformation and an opportunity to improve growth, relationships with customers, and productivity. The transformation must be rooted in business value, and the results have to be measurable.”
All well and good, but no matter how strategic and transformative the benefits of AI, the question of where to start is an important one to answer. The key challenge to getting a meaningful return on investment from agentic AI is understanding which business process the technology can optimize. For example, there are some specific tasks (e.g., looking up an in-force policy) that are done by multiple departments such as claims, contact center, and so on. The same agent could add value across all of these. Identification and quantification of those opportunities is the challenge.
The Challenge: AI without focus can become expensive ‘innovation theater’
Many companies leap into AI pilots driven by hype, only to discover the chosen use case doesn’t generate the savings – or revenue increases – that are expected. Nine times out of ten, that disconnect comes from not fully understanding which processes and tasks take up the most time and effort in your operation, how often they occur, who does them, and how much they cost. Unfortunately, the increasing pressures on time to market and cost mean that insurers who don’t have a good enough understanding of the unit costs associated with key tasks will lose out to those that do. Overcoming this challenge involves knowing exactly where and how work gets done in your organization, so you can identify where agentic AI will replace or enhance work in a way that generates real business value.
Decision Intelligence: real-time insights into your work, people, and processes
A highly focused, data-driven, systematic approach is the best way to get started.
Pinpoint High-Impact Opportunities
First, gain a comprehensive view across systems and organizations to best understand the best opportunities for agentic AI:
Identify Repetitive, Rules-Based Tasks
- Identify repetitive, rules-based tasks in claims, policy servicing, or underwriting that are ripe for AI delegation.
- Spot process variations where AI could bring standardization and speed.
- Think beyond the work AI is doing. What high-value and fulfilling tasks will AI enable your people to focus on?
Quantify the Potential Savings
- Use actual work volume, handling time, and error rate data to model how much capacity, cost, and turnaround time AI could save.
- Leverage the data to understand the potential downstream implications of AI agents. If they are completing tasks faster or with fewer errors, do other workflows speed up?
- Don’t forget to quantify the benefits of the work that your humans are now able to do with their time – or the projected benefits of their increased satisfaction levels.
Understand Human + Digital Workforce Interactions
- See where human judgment adds value, so AI complements areas where human judgement is required.
- Reduce risk by keeping humans in the loop where regulatory or customer sensitivity is high.
From data to action: AI readiness in insurance
The insights above enable you to prioritize AI investments based on ROI, not guesswork. With this data-driven foundation, you can develop a targeted AI roadmap:
- Start small: Deploy agentic AI in low-risk, high-volume processes like data entry and validation for FNOL (First Notice of Loss).
- Scale smart: Expand AI into adjacent processes based on proven time/cost savings.
- Sustain success: Continuously monitor AI performance and its impact on service levels, compliance, unit costs, and staff capacity.
The Results: Time, money, and capacity freed up
With this focus, insurers typically see:
- Faster processing times — reducing claims cycles and improving customer satisfaction.
- Lower operational costs — by cutting manual work hours and error rework.
- Freed capacity — enabling staff to focus on high-value work like complex claims or proactive customer engagement.
Agentic AI has the potential to transform insurance operations — but only if deployed where it will truly make a difference.
It’s one of the reasons I’m excited to be involved with ActiveOps right now. The key to delivering on the vision I’ve described is data – real-time, trustworthy insights from across your entire operation. ActiveOps’ ability to deliver an operational ‘x-ray’ across claims, underwriting, service, and other systems is in many ways unique, and it has potential to deliver visibility into every work type, process flow, and productivity trend across both people and digital workers.
Many organizations are currently deploying Decision Intelligence with a range of objectives in mind – managing growth and professionalizing operations to name a few – and Agentic AI looks like another great use case. It doesn’t guarantee success – but it does guarantee you’re making informed decisions, giving you the best chance of success.
About the author
Stephen Siegel has worked in insurance technology and operations for over 25 years. He designed and managed the implementation of one of the first successful AI systems in insurance. He has worked in virtually all segments of insurance, spanning distribution management, new business, underwriting, claims, service – and more.
Related blog posts
