In workers’ compensation, AI is moving from proof point to power shift.

Building on findings from our new Guidewire-PwC report, Reimagining Workers’ Compensation in the Age of AI, this three-part series has showcased why the case for change is accelerating and where early results are starting to surface. In this final installment, the real significance of AI in workers’ compensation isn’t that it can improve key tasks. It’s that those early wins point to five broader forces that, taken together, could completely redefine this line through 2030.

AI in Workers’ Compensation: No Longer Just an Automation Story

Much of the conversation around AI in insurance defaults to process automation and cost savings. That still matters, of course. But in workers’ compensation, the question is no longer whether AI can help summarize a document, surface a recommendation, or cut handling time. It’s what happens when carriers start using AI not just to speed up work, but to redesign how the line actually operates.

That distinction matters because the five forces we identified in our study are not discrete trends. They’re five ways the line begins to change once better understanding, better timing, and better decision support start showing up in core workflows—with material results that include the following:

Recovery Gets Personal

For most of its history, workers’ compensation claims management has been built around institutional processes. That made sense. Carriers organized around tools, staffing models, and the controls they had at their disposal. AI creates the conditions for something far more worker-centric.

An injured party has traditionally had to manually complete and submit notification and details of an injury to their employer and insurer. AI combines simplified, digital first-report-of-injury (FROI) processes with enriched data from third parties and the Internet of Things (IoT) to reduce the time it takes to report an injury.

Where standardized return-to-work (RTW) plans have long aligned strictly to common medical coding and loss categories, AI customizes RTW plans based on an accurate assessment of the individual’s specific injuries.

When injured workers can get more personalized support, the model starts to shift. Claims handling becomes less about managing steps and more about managing recovery. Rather than reducing discipline, this makes the process more responsive to the person moving through it.

Workflows Get Smarter

Workers’ compensation carriers are under pressure to manage higher caseloads with greater speed, precision, and personalization. AI can help by reducing the administrative drag inside the middle office and improving the flow of decisions across intake, triage, injury analysis, and proactive claims support.

“We’re targeting the invisible moments that might make the biggest difference to our staff and the people we serve,” said Gavin Pokan, Chief Recovery and Care Officer at Ontario Canada’s Workplace Safety and Insurance Board (WSIB), in a recent episode of my InsurTalk podcast. “The net effect is fewer hours lost to high-friction tasks so our people can spend more time with injured workers resolving issues and tailoring support. That’s the part of the job that actually moves outcomes.”

This is where the distinction between point-based tools and enterprise value comes into play. A summary alone isn’t transformative. One produced by AI based on insights from unstructured medical records, injury narratives, and historical outcomes that then embeds guidance directly into workflows is. A recommendation alone isn’t transformative. A recommendation that takes early warning signs into account and triggers an intervention to prevent or mitigate attorney involvement is.

In our study, we found that forward-looking carriers are establishing AI centers of excellence to enable sustained, enterprise-wide adoption and impact on core processes. Because the opportunity is about a whole lot more than just adding intelligence to existing workflows. It’s about making the workflows themselves more adaptive, more connected, and more decision-rich.

Expertise Stops Walking Out the Door

The Bureau of Labor Statistics projects that in the US, there will be 21,600 openings a year for claims adjusters, appraisers, examiners, and investigators through 2034, with all of these openings expected to result from seasoned professionals retiring or otherwise leaving the industry. The same trends are taking hold in markets like the UK, where more than a quarter of all insurance professionals are over 50.

What some call a “silver tsunami” could decimate decades of institutional knowledge and eliminate what WSIB’s Pokan calls the age-old practice of talking over the baffle to seek guidance from more experienced coworkers.

AI offers a path forward. Using AI technologies like Guidewire ProNavigator, for example, workers’ compensation carriers can capture this expert knowledge and instantly embed it across underwriting, distribution, and claims management, reducing reliance on lengthy apprenticeships without eroding decision accuracy or service quality. As I outlined in a recent post, I believe AI will help the next generation move faster and more confidently than the last, not because they know more, but because they start with expert knowledge built directly into the systems they use. As a result, the worst impacts of the retirement cliff workers’ compensation faces will be blunted, even as decision accuracy, velocity, and consistency improve.

Loss Gets Lost

The fourth AI force reshaping workers’ compensation is a shift from loss transfer and post-injury recovery toward proactive risk management, where carriers play a more active role in prevention and early intervention.

Workers’ compensation has always prioritized health and safety. What AI changes is the ability to analyze far more unstructured information, from incident reports to maintenance logs and clinical patterns, in order to identify hidden vulnerabilities before injuries occur.

Carriers operating a cloud-based insurance platform that integrates data from remote technologies like those from Kinetic can reduce injury rates by up to 58% while cutting claims costs by up to 54%. CopperPoint Insurance, for example, leverages its infrastructure to integrate AI with on-site video monitoring to spot unsafe work activities and other hazards to prevent injuries from happening in the first place.

When injuries do occur, AI enables early intervention by accelerating the claims process, tempering the impact of workplace injuries. Quick response systems, coupled with coordinated medical care and modified duty programs, also help injured workers recover faster and return to productivity sooner, reducing overall claim costs.

Platforms Eat Point Solutions

Workers’ compensation is too complex for any one carrier to build out every critical AI capability alone. Nor is that even necessary. Today, a considerable percentage of the innovation in this market is coming from specialized ecosystem partners solving specific problems: bill review, care management, claims guidance, predictive analytics, reporting, and more. The strategic question isn’t whether to use outside innovation. It’s how to turn it into an integrated operating advantage rather than managing and maintaining a vast array of disconnected point solutions.

I realize I’m biased, but by now it’s clear that a modern, API-first core and data platform with a well-defined AI governance framework is essential to quickly innovate new capabilities and services to better meet customer needs.

Using Guidewire as an example, our insurtech incubator program and industry-leading solution ecosystem span claims reporting/FROI, claims management, care management, medical intelligence, payment and recovery, documentation and communication, and much more. But whatever path they choose, carriers should look for an AI-first platform that aligns with their business goals, offers pre-vetted solutions that easily integrate to the core system, and delivers a unified, technology-empowered approach to managing risk and serving customers.

The Leadership Agenda Now

For workers’ compensation leaders, the path forward isn’t mysterious, but it is demanding. As I’ve seen firsthand, that means they must:

• Start with the highest-value AI use cases—especially ones that touch both economics and outcomes
• Move toward a flexible core architecture that supports AI agents and applications at scale
• Implement robust governance, security, and observability to ensure safe, compliant AI usage
• Leverage connected solution and data ecosystems selectively and strategically
• Commit to scaling a small number of high-value use cases rather than admiring a long list of pilots

That’s the real message of this series and the report behind it. As better, faster timing, judgment, and coordination grow mission-critical, AI is becoming an indispensable technology. The carriers who understand that will be best positioned to lead whatever comes next as workers’ compensation’s AI revolution continues.

Read the Report, Join the Conversation

Be sure to read the full report, which I co-authored with Imran Ilyas, Partner, PwC; Oliver Winkenbach, GM, Data Applications, Guidewire; and Matthew Wolff, former partner, PwC. You can also listen to a recent episode of my InsurTalk podcast, where my collaborators and I discuss key findings from the report, below: