Excitement about adopting artificial intelligence technology is palpable in the insurance industry, but the reality is that while most carriers use artificial intelligence technology, few use AI at more than a small scale— and the vast majority of adjusters say need AI needs human oversight.
A new report out from Sedgwick, Future-ready property claims: Leveraging technology and AI for a strategic advantage, estimates that between 58% and 82% of insurers use AI tools in their operations, however just 12% of say they have fully mature AI capabilities, and only 7% say they have achieved scalable AI success.
The report paints a picture of accelerating AI adoption in the industry, but with carriers at vastly different points in the process: some are still dabbling and experimenting while others scaling AI across operations.
Related: Carriers See Higher Claims Severity Amid Medical, Social Inflation and Growth in AI‑Generated Fraud
David Guaragna, senior vice president, Sedgwick Property Americas, explained the dichotomy.
“People look at insurance claims as a monolith, and if you think about it, different carriers specialize in different lines, and there’s just so much complexity across the insurance space,” Guaragna said. “And if you look at that different complexity, the data associated with those different lines is extremely challenging and it’s approached different ways. So, to think that everyone’s going to move at the same speed and the different processes with which they collect that data and the different systems involved—whether it be the property space requiring estimates, in the auto space requiring different estimating platforms, they’re relying on different partners and different systems—I do think that they’re going to move at different speeds based on the need for that data.”
The value of AI in insurance is expected to reach nearly $80 billion by 2032, up from just $10 billion in 2025, according to the report, which cites a Moody’s analysis of ways that AI is transforming the industry. Despite the significant investments being made, fragmentation is limiting AI’s impact in such a way that any value being realized is from individual claim steps, according to the Sedgwick report.
Related: AIG’s Zaffino: Outcomes From AI Use Went From ‘Aspirational’ to ‘Beyond Expectations’
Nearly two-thirds of carriers say there’s a gap between their AI vision and their reality. As a result, carriers “have a fragmented technology experience” in which different tools and vendors support different parts of the claims process, according to the report.
“With so many tools involved in the claims process, carriers’ data is often inconsistent, incomplete or siloed across systems, which weakens AI outputs and decisions,” the report states.
A new analysis from Goldman Sachs shows a divide between “boardroom hype and macroeconomic reality,” which was covered recently in Fortune. “We still do not find a meaningful relationship between productivity and AI adoption at the economy-wide level,” stated senior U.S. economist Ronnie Walker in an analysis of fourth quarter earnings.
According to the Sedgwick report, using AI to handle low-severity claims has led to 80% faster processing times for some carriers.
Related: FM Using AI to Elevate Claims to Deliver More Than Just Cost Savings
Speed and cutting costs appears to behind the adoption of an AI claim assistant now taking auto damage claims calls at Travelers. The service, developed using OpenAI model capabilities and APIs, uses advanced language and speech recognition technologies. Travelers reported in February that the capability is initially being used with customers who are calling to file an auto damage claim.
Tim Parker, a long-time claims professional and host of The Adjusters Diaries podcast, is consulting for Google’s Gemini AI information platform.
Based on his work with Gemini, he’s seeing a prime use of AI as a source of information for claims professionals.
“If you look at the types of questions that are coming in, they’re all claims-related questions,” Parker said.
Frequent questions center on users who say they are “running a loss report for XYZ company,” and they need to know how much a carrier will pay for on certain reconstruction costs, or other basic information about property damage.
“What I do is I look at the questions that stumped the AI and I provide the correct answers, so people that are building out the code can begin to understand it,” Parker said. “We see the insurance company deploy AI through Apple and Android platforms that will allow the insurer to, on specific losses, write their own estimate and if there’s a question, they’ll send not an adjuster but they’ll send a contractor out to review it and then send it up through an (AI) program.”
Companies are using tools like scrubbing platforms, in which AI can look at the information and tell the company whether they need to send the claim to a human to make a judgment call, Parker said.
The report also shows that 75% of claims professionals believe AI needs human oversight.
Guaragna believes the report indicates what many in the professional already know—people are needed in all but the simplest of claims.
“AI will enable adjusters to be more efficient. It’s a tool to be more efficient and more productive, but we will need the human touch, especially in claims where you’ll always need that empathy with policyholders,” Guaragna said. “A claim is always a challenge for a claimant no matter what type of claim it is. Whether it be a personal lines property claim or workers’ compensation claim, there needs to be a human element involved with the understanding that there is a human-to-human need to understand that someone’s going through a challenge in that process.”
Other findings in the report include:
- 82% of carriers use AI for routine tasks (data extraction, automated customer interactions)
- Intake automation has reduced average claim processing times from 10 days to 36 hours.
