Since the start of this decade, insurance has transformed dramatically. Technology is evolving at such a rapid pace that the sector could look vastly different in the next few years. With 2030 edging ever closer, FinTech Global spoke to leaders from across the sector to get their thoughts on what the insurance sector will look like in the new decade.
The three biggest shifts
Dan Simmons, managing director and founder of Quensus, sees the biggest transformation being the move from retrospective indemnity to continuous, autonomous asset protection. A core driver of this will be the growing climate hostility, which is causing greater and more frequent damage that is forcing insurers to pay out claims at unprecedented levels. To counter the risk, insurers will need to move a “pay to rebuild” mentality and to a model of “actively protecting.”
At the centre of this movement will be the adoption of agentic AI, he explained. The technology will be able to continuously and automatically monitor systems, taking necessary action to prevent an incident spiralling into a major claim.
He said, “Rather than relying on human-in-the-loop notifications, advanced machine learning models will identify abnormal consumption patterns and autonomously trigger mechanical shut-off valves to isolate water supplies instantly, effectively eliminating the “repair lag” and driving attritional loss severity toward zero.”
In a similar line of thinking, Matthew Twist, vice president of sales, EMEA at Earnix, believes the biggest difference between insurance in 2026 and in 2030 will be the speed at which insurers can act.
The sense of immediacy has taken hold of the world. Being always connected to the internet, people have grown accustomed to not having to wait. Whether it is watching something on-demand, ordering an item for next-day, or even same-day delivery, real-time access to finances, connecting with people across the world, or the ability to get answers to questions in nanoseconds, people no longer need to wait. This has shaped what people demand from service providers and insurance is no different. Waiting weeks or months for a claim to be processed risks customers seeking other providers at renewal dates.
Twist explained, “Risk, customer behaviour and market conditions are changing so quickly that the useful life of an insurance decision is getting shorter. Insurers need to recognise change and respond with confidence.”
Agentic AI again finds itself at the heart of the solution. In just a matter of years, AI has advanced from a technology that can help people analyse information into a specialised agent that can work autonomously directly within workflows with defined permissions and human oversight. This is helping insurance teams to become faster in leveraging intelligence.
Twist noted, “Intelligence from underwriting will increasingly inform pricing, while customer insights will shape what happens next across the business.”
The time savings insurers can achieve through agentic AI vary across the value chain. A recent report from McKinsey claimed that productivity could improve by between 10% and 90% across various stages of insurance.
Melanie Hayes, co-founder at KYND, offered a different perspective. For her, the biggest shift in insurance will be the move from declared risk to observed risk. She explained, “Commercial insurance has always relied heavily on what a business tells you about itself, and for a long time that worked well enough. What’s changed is the pace. The technology a business runs today can look quite different in a week, and AI has sped that up considerably.”
Today, a company can adopt a chatbot or AI agent tool within an afternoon. This simple change can completely alter its risk posture, but it is unlikely to be something they would even consider mentioning at the policy renewal. In areas like cyber and technology errors and omissions this will have the greatest impact as the technology is part of the exposure.
Simply adding an AI tool might not seem like a major change to a company’s risk posture, but it can have major knock-on effects. For instance, if the business lacks proper governance frameworks, the technology could make a mistake that leads to significant damage. Alternatively, if a third-party provider suffers an outage, large workflows of a business can instantly become impacted, or a cybersecurity failure can leave firms exposed.
But forgetting to declare a new AI tool is not the only reason Hayes sees the market moving to observed risk. The other contributing factor is the desire for speed and simplicity. Hayes added, “Munich Re expects the global cyber market to more than double between 2024 and 2030, and firms, quite understandably, want buying cover to be quick and simple. So I think we’ll see more of the picture built from what can be observed from the outside, with the business confirming the details rather than filling in pages of questions. Easier for the buyer, and a clearer view for everyone else.”
This will also lead to a change in where portfolio thinking happens. Hayes explained, “Accumulation has typically been something you look at once risks are on the book. The CrowdStrike outage in July 2024 was a real moment for the market, because it showed how one technology provider could link thousands of businesses that otherwise had nothing in common.”
Caused by a logic error in a sensor configuration update led to a seismic outage that impacted over 8 million Microsoft Windows computers that were unable to properly restart.
As more and more businesses rely on the same cloud platforms, identity providers and AI models, and as a result the shared threats are multiplying. The risk net continues to grow and insurers will need to account for this. Hayes added, “By 2030, I’d expect that view to sit alongside the individual risk at the point of quote. (And with quantum computing edging closer to breaking today’s encryption, and AI speeding up how quickly weaknesses are found, the next few years may bring shared exposures none of us can fully picture yet.)”
Where will the biggest change be internally?
Without a doubt, the area of the traditional insurance value chain that will see the biggest change by 2030 will be manual workloads. All three contributors agreed that this type of work will increasingly disappear or see a significant reduction.
Hayes explained that manual back-and-forth wastes a lot of time within teams. Staff are still forced chasing, re-keying and checking information, despite much of this being available directly from a business’s digital footprint. By using technology to automate this, firms can instantly save multiple working hours for their staff. “For small and mid-sized business cyber cover in particular, that opens the door to quoting and binding with very little effort from the buyer or the broker, and to routine renewals being largely automated.”
In the same vein, Twist believes manual handoffs to disappear. Moving data between systems, checking inputs, preparing routine analyses and searching for product information are perfect tasks for AI agents to take control of and free up time for practitioners. Some straightforward workflows could also be among the workloads that are replaced, with clear boundaries in place to ensure the output is managed. Humans would instead spend their time handling complex underwriting, exceptions and decisions where there are significant human consequences.
He said, “By 2030, I expect professionals to spend much more time exercising judgement, improving products and helping customers. Connecting these capabilities within existing technology environments will be essential to making that transition practical.”
Focusing on the commercial and personal property space, Simmons sees traditional manual underwriting coming to an end. Instead, deep learning algorithms will collate real-time IoT and API data to execute underwriting decisions within seconds. This technology will also change the conventional claims process, with parametric models and smart contracts transforming the work.
He said, “When an IoT flow meter detects an anomaly, it will autonomously isolate the risk, log the event on a distributed ledger, dispatch a repair technician, and instantly settle the attritional claim without requiring a human loss adjuster’s intervention.”
Underestimating the future
With the AI in insurance market projected to be worth $154bn by 2034, it is safe to say the insurance sector is not underestimating the impact AI will likely have. However, there are some areas that are flying under the radar more than they should.
For Twist, insurers are underestimating the lasting impact of the technology and supplier choices they are making today. These decisions are going to have a big influence on performance in 2030, but also how quickly insurers can respond to evolving regulatory landscapes.
He said, “Insurers should be asking whether suppliers understand insurance decisions, can demonstrate traceability and accountability, and can support changing rules without forcing a wholesale rebuild. Governance needs to work in everyday operations.”
Elsewhere, Twist also sees firms potentially underestimating how much value comes from connecting decisions. Regardless of how great the model is, its impact will be diminished if the insights arrive too late or cannot inform actions needed elsewhere. “The strongest insurers in 2030 will combine intelligence with control. Choosing trusted suppliers and adaptable foundations now will give them the freedom to move faster, serve customers better and turn agentic AI into sustained business performance, not an oversight risk.”
Simmons, on the other hand, believes the industry is underestimating the disruptive impact of autonomous physical mitigation. He pointed to IoT, stating that the industry currently perceives it as a descriptive tool that sends passive alerts to human operators. However, agentic AI is turning them from passive to proactive tools that are capable of minimising damage and claims.
He said, “Agentic AI empowers systems to physically intervene and neutralise threats in real time without human oversight. When you eliminate the human delay in responding to a water leak, you essentially eradicate the attritional loss entirely. This absolute mitigation will disrupt current actuarial assumptions, rendering historical loss ratio models obsolete and fundamentally altering capital reserving requirements.”
Finally, Hayes believes the industry is underestimating how much buyers’ expectations are about to change. She explained that most businesses run their banking, payroll and accounting through tools that give them instant, clear answers and many are using AI assistants for everyday decisions. It is almost certain they will want the same from their insurance, she stated.
Hayes said, “By 2030, I think a cyber or technology policy will be judged not only on what it pays out, but on what it tells the business along the way. Many cyber losses trace back to issues that were visible well before anything went wrong: an unpatched system, a poorly configured email domain, a supplier dependency nobody had mapped.
“Threats keep advancing, and AI is helping attackers write more convincing phishing and find weaknesses faster, yet the way in is still very often through those basics. Keeping them secure matters as much as ever. When a business hears about gaps like these early, in plain language and specific to its own set-up, it tends to act. For brokers, that’s a real opportunity, because a conversation that starts with insight rather than price is a very different conversation.”
