If AI Is So Efficient, Why Aren’t Insurers Seeing the Savings?

There's a lot of AI-hype. Where are the results?
By: | March 12, 2026
Topics: Risk Insider

I’ve been struck by a growing disconnect between what recent research suggests and what I’m hearing in conversations with claims leaders.

On one hand, industry analyses and operational benchmarks suggest that artificial intelligence can increase the speed and efficiency of key claims activities. FNOL handling becomes faster, document processing accelerates, and elements of damage assessment move more quickly when workflows are redesigned around automation. In the right conditions, these improvements are real and visible.

At the same time, research from MIT and Deloitte suggests that the vast majority of organizations across industries have yet to see measurable ROI from their AI investments. Depending on the study, only 5–15% of firms report meaningful bottom-line impact so far.

Some of this gap comes from grouping different forms of AI together. Certain organizations are seeing returns from narrower generative AI use cases, while far fewer report ROI from more complex agentic systems designed to automate end-to-end processes. These approaches offer greater potential, but they also require deeper integration, workflow redesign, and time.

Even when the technology works as promised, efficiency does not always translate into savings. AI can make claims faster while organizations still fail to see ROI. Both realities can coexist. The question is why.

Many of the gains being cited reflect process-level improvements. They show what is possible when individual steps such as FNOL, document intake, triage, and coverage checks are automated on their own. Many of these gains are real and repeatable today. What often goes unmeasured is what happens when those automated steps meet existing operating models. In most cases, the surrounding process absorbs the efficiency rather than eliminating cost.

This shows up in day-to-day operations. Claims organizations still re-key AI outputs into legacy systems. Automated decisions are routed back to adjusters for review, often out of caution. Manual workflows continue to run in parallel for compliance or comfort. The result is AI layered on top of the old workflow, not AI replacing it.

As a result, cycle times tend to fall faster than costs do.

This is the gap claims leaders are navigating. They are shown efficiency gains but held accountable for expense ratios and outcomes. The question shifts from “is AI working?” to “what needs to change around it?”

The organizations reporting more meaningful gains focus on that second question. Rather than task automation alone, they redesign workflows from start to finish. They treat AI as part of their operating infrastructure, not just a productivity aid. That is when external adjusting spend begins to decline, severity leakage decreases, and straight-through processing becomes more than a goal.

For risk managers and CFOs, the takeaway is not that AI fails to deliver value. It is that potential savings and captured savings rarely line up on their own. Until process ownership, system integration, and measurement evolve alongside the technology, ROI can continue to feel elusive.

If a process truly takes minutes instead of hours, what no longer needs to exist around it? That is where the savings actually show up.

Sean G. Eldridge is the Co-founder and CEO of Gain Life, a venture-backed insurance technology company that helps individuals and organizations return to health, work, productivity, and financial wellbeing. Prior to Gain Life, Sean led a private equity-backed roll-up in the disaster restoration space and held leadership roles at Johnson & Johnson, Procter & Gamble, and Weight Watchers, launching new products and services that leverage the power of behavioral science and technology. Sean earned his B.S. in Management Information Systems from Rochester Institute of Technology and MBA from Harvard Business School. He resides in Cambridge, MA with his wife, son, and corgi.

More from Risk & Insurance