This year, I joined Sprout.ai as a strategic advisor and member of the Growth Advisory Board because I was excited by the opportunities that AI offers for the insurance market and impressed by what I had seen the company achieve. It was clear that there was a very engaged and skilled team that had developed some great solutions for an impressive client base.
I certainly feel I could add value, which is very important to me in deciding who I work with. I knew that I could bring my many years of insurance intrapreneurship experience to help shape, refine and ensure that the Sprout.ai’s solutions could address the real transformation challenges that the insurance industry faces, and thus support the growth of the business.
In terms of pleasant surprises, I hadn’t appreciated the extent of Sprout.ai’s client footprint geographically or the wide range of lines of business for which solutions had already been successfully deployed. Equally, I hadn’t fully understood just how many different languages the team had delivered these solutions in. It is incredibly impressive to me, for instance, that they have developed the capability to assimilate and correctly determine the intent of handwritten Japanese Kanji and Kana.
Given the range of Sprout.ai propositions that are already delivering in multiple lines of business across a range of culturally diverse organisations, the potential for the future is very exciting. There are huge opportunities for Sprout.ai to cross-fertilise their solutions within companies, along the insurance value chain, and between different markets and countries.
At the same time, Sprout has yet to have delivered at scale in some of the larger insurance markets and sectors, offering even more opportunities to expand the business. Ultimately, I joined Sprout.ai because I’m excited about the team, I’m excited about the technology, and I’m excited to continue learning and developing myself through working with them. Most of all, I’m thrilled to help further catalyse the accelerated transformation that the insurance industry so sorely needs.
Here are a few thoughts on what that journey could look like.
The insurance market is awash with stories of how AI can enhance efficiency and provide slicker ways to interact with customers. However, what has not been fully grasped is the opportunity to achieve more impactful results for the top and bottom-lines that go beyond these more obvious benefits.
Firstly, AI has the potential to reduce claims and underwriting leakage, ensuring that the right cover is provided at the right premium and that the cost of settling claims is no more, or less, than it should be. For instance, at least 75% of insurers total outgoings is spent in settling claims, so even a small percentage saving on indemnity costs can move the business dial far more significantly than the same percentage saving on expenses.
With carriers and others in the insurance ecosystem facing significant inflation challenges in highly competitive areas such as auto-insurance and commercial rates in many lines already softening, the ability to more effectively price coverage and control claims costs will accelerate as a significant competitive and loss ratio advantage.
Secondly there is still too little emphasis on educating and engaging the real underwriting and claims technical and process experts in Ai-enabled transformation. By bringing-in ‘at the sharp end’ people to identify the mundane, long-winded, or complex but low-value tasks, brings a greater ownership of change and can accelerate the transformation process. In addition, this not only empowers employees but also allows them to determine where the freed-up resource can be best focused to add the most value – whether that’s investing in personal development, building more proactive relationships with customers, or managing workloads and tackling backlogs more effectively. The full advantages of a positive, engaged and motivated workforce are difficult to quantify but we all know the huge benefit this brings for customers, partners and the business.
Real-world lessons in AI deployment
In terms of effective AI applications, I’ve encountered some standout examples. For example, I’ve seen large language model solutions help uncover millions of dollars in missed reinsurance claims. A relatively short and inexpensive development, led jointly by claims and AI experts, delivered a solution that flagged incorrectly coded catastrophe claims that missed from reinsurance recoveries. This, to me, demonstrates that impactful outcomes need not require costly or lengthy implementations and that targeted, agile applications of AI can deliver very impressive results even in an infrastructure under-pinned by legacy systems.
Why insurers hesitate to start using AI
Despite these successes, reticence remains among insurers when it comes to adopting AI. One of the main concerns is the fear that AI applications might ‘contaminate’ data or processes beyond their intended scope. Data integrity, personal data protection, and security risks also loom large, as does the perception that AI could introduce biases into decision-making or amplify existing process flaws.
Legacy systems present another hurdle. From my discussions with claims leaders at a Camelot-facilitated Sprout.ai roundtable, it’s clear that some insurers are pursuing a re-platforming strategy before thinking about where AI offers incremental opportunities, while others seek more agile solutions that can work with existing infrastructures. Regardless of the approach, resource prioritisation is a practical challenge. Business and Operations teams already stretched thin with customer service, backlog and ‘keeping-the-lights on’ project demands often lack the capacity to explore AI opportunities.
Then there’s the issue of ownership. Without shared buy-in from leadership and operational teams across different functions, AI initiatives can become mired in internal politics or misaligned to the business priorities. Add to this to the tendency for some COOs to favour the safe path of partnering with large, well-known companies rather than finding ways to work with smaller, more innovative and agile players, and it’s easy to see how due-diligence and on-boarding processes can drag on, risking both a loss of momentum and wasted effort for everyone if sponsors change roles or priorities shift.
Starting small, thinking big
For leaders and decision-makers hesitant or bewildered as to where to start with AI, my advice is to start with smaller, more easily deployed, ‘quick-win’ projects that demonstrate both clear, quantifiable benefits and help build confidence. Automating mundane tasks can deliver streamlined outcomes quickly, reduce errors, and improve resource efficiency. Success in these smaller initiatives can then act as a springboard and build momentum for broader, larger scale solutions.
Another key factor is peer advocacy. Hearing a fellow business leader share their positive experiences and advocate for AI carries more weight and can act as a far more effective change enabler than a message from someone who is seen as disconnected from the reality of day-to-day business challenges.
Addressing ethical concerns
As touched on previously, ethical concerns, such as the potential for bias to be developed or amplified by AI-driven decision-making, are another area insurers must address. Avoiding these actual or perceived pitfalls requires early engagement of those accountable for compliance and risk management during solution design and effective testing to ensure consistent outcomes and address concerns. Ultimately, incorporating human checks and balances builds confidence in AI in trapping any grey-area decisions. This enables algorithms to be refined to ensure the proportion of tasks where AI can be relied-upon to provide a consistency of the desired outcome can be increased.
It’s worth remembering that errors in processing aren’t new. We have always recognised that human and system errors occur and have built processes and controls to avoid or mitigate the risks that arise long before AI came on the scene. The fact is, well scoped, developed and deployed AI solutions delivered by a combination of those who know the business best and those with the best technology skills will be more effective in reducing incorrect outcomes than any simple process that relies on human intervention.
Tipping points and transformations
Looking ahead, I believe the tipping point for widespread AI adoption will come when the optimum balance is achieved between those who see only the risks and those who are energised and excited by the opportunities. As use grows, so will confidence and trust. During this period, some organisations may pull back, but others will refine their approaches and be the leaders of the change our industry needs.
The key to accelerating adoption is recognising AI for what it truly is: a tool to supplement and support, not replace, people and expertise. Practical deployments that deliver measurable benefits will help shift perceptions, enabling leaders and teams to see AI as a partner that improves outcomes and frees up energy to focus on higher-value activities. When this happens, insurance AI solutions will no longer be perceived as a threat that some see them as now but as an enabler of more high quality and consistent decisions and services, as well as a more positive work environment. Ultimately, the next ten years will determine which organisations have the purpose-driven leadership and determination to shape the future insurance success stories and which will become the brands of the past.
Where AI fits best
Opportunities for AI in insurance exist across many areas. High-volume, low-complexity processes like gadget or travel insurance claims are the most obvious and, for many companies, the starting point on their AI transformation. Yet there are still so many unexplored opportunities in this area. For example, Life Insurance remains a largely traditional and paper-driven environment which has seen limited innovation and therefore represents fertile ground for AI-driven transformation.
Boiling the ocean to deliver the ultimate end-to-end transformation is not likely to be the best strategy, either. AI is very effective at addressing subcomponents of processes, like assessing policyholder eligibility or streamlining subrogation recoveries.
This then neatly brings me on to the myth that there are limited opportunities for AI in complex insurance areas, such as commercial business. These are, in fact, often target-rich environments for astute innovators and agents of change. Processes in complex insurance areas, when broken-down, often reflect the same characteristics as those considered as less complex. For example, most auto claims are processed in the same way for retail and commercial customers.
Equally, while an overall process may be complex, specific sub-components may offer great opportunities for AI based solutions, examples being elements of the workers’ compensation value chain, such as medical invoice checking and payment processing. AI-driven improvements can also add significant value in, for example, summarising lengthy expert reports or ensuring the consistency in the terms and conditions of international policies issued in multiple territories.
Ultimately, AI adoption will scale when organisations move beyond the hype and start delivering tangible, practical outcomes. Insurers must embrace AI as an enabler of better customer experience, operational efficiency, outcome effectiveness, and an enhanced working environment. With the right focus and collaboration, AI has the potential to transform not just processes but the industry itself.
With over 30 years of experience in the insurance industry, Ian Thompson brings vast expertise in claims management, digital transformation, and data-driven innovation. His recent role as Group Chief Claims Officer at Zurich saw accountable for leading a team of 8,500 across 200 territories, handling 7.5 million claims annually.
