Spotlight: AI drive revs up the motor race

In the fiercely contested world of motor insurance, timely and intelligent use of data is helping firms maintain a competitive edge. Laura Miller reports.

When stock markets opened on Monday 27 January 2025, global tech stocks fell sharply as the arrival of DeepSeek’s R1 open-source model rocked the artificial intelligence (AI) industry. It also opened the way for financial services firms, including motor insurers, to pursue new AI approaches.

“R1’s launch underlines both that AI innovation is still accelerating, and that predictions that smaller language models would be a compelling option for organisations was correct,” suggested one data and generative AI expert Insurance Post spoke to.

For example, smaller models give firms the opportunity to leverage and curate their own training datasets, due to the lower data requirements needed to train smaller language models.

The ability for organisations to use their own training data – rather than relying on hyperscaler datasets – can enable financial services firms to develop more fine-tuned and relevant models, and improve competitiveness.

This is particularly valuable for institutions who may need to accommodate diverse customer and compliance needs across multiple sectors and countries, like motor insurers.

Policyholder information, claims history, underwriting details, as well as external data like demographics, credit scores and location data can be used to drive machine learning (ML) models for pricing, underwriting and claims analysis.

Likewise this data can be leveraged to provide comprehensive risk assessments and improve decision making, with the aim of enhancing pricing, claims and underwriting performance, and the holy grail of automatically adjusting to changing circumstances and market events in real time.

Motor insurers say they are already using some AI tools, with limitations. Allianz, for example, uses AI and automation to respond to emerging trends quickly in all functions like fraud, adverse performance and customer journey.

Henry Topham, managing director of retail insurance at Allianz UK, says, “While AI/ML can help organise and display the key trends, and automated reports have removed a lot of the manual work…there is still a requirement for intervention to explain the ‘why’ where we rely on the expertise and experience in pricing and underwriting”.

Data acceleration

The global insurance data analytics market was estimated at $13.29 billion in 2025, according to Mordor Intelligence. It is expected to reach $27.8 billion in the next five years. More than eight in 10 (86%) insurers are creating insurance data analytics systems to provide the most accurate predictions of big data reports.

Data analytics enable unprecedented creativity across all product categories and corporate functions. With the rise in competition in the insurance sector, the need for analytics solutions is rising to sustain stiff competition across the global market. Joe Farrar, pricing director at AXA Retail, explains how that works at his firm.

“Using the opportunity to capture and analyse more data allows insurers to move away from traditional motor insurance policies to propositions with fair pricing for customers’ individual circumstances,” he says.

“We know that cost is one of the main factors that customers consider when purchasing a policy, so it’s important that we offer products and services with flexible price points, which is something that data can help shape.”

Fighting rising levels of motor fraud is another benefit to insurers of properly harnessing data and analytics. In 2023 alone, 84,400 fraudulent claims worth £1.1 billion were detected by the Association of British Insurers in the UK, a 16% increase in the 

This data insight can help to ensure fair pricing for customers while safeguarding our resources against fraudulent claims activity.

Joe Farrar, AXA Retail

number of detected claims compared to the previous year.

For motor insurers there exists, Axa’s Farrar says, “huge benefit” in being able to quickly analyse vulnerabilities to identify potential risks, “and insurers can leverage AI to detect and combat motor insurance fraud specifically”.

He says: “As fraudsters become more sophisticated in their methods, it is becoming increasingly difficult to spot fraud trends early.

“We regularly monitor claims against our expectations to ensure trends are identified and prices reflect our latest view of risk. This data insight can help to ensure fair pricing for customers while safeguarding our resources against fraudulent claims activity.”

Success in this and other areas necessitates insurers’ underwriting and claims departments working more closely together so underwriters get a better read from claims data. And there is a lot of data to work through.

Over the course of 2024 UK insurers dealt with 2.4 million motor insurance claims and paid out a record £11.7 billion, according to the latest data from the ABI. Increases in theft and repair costs continued to impact premiums, it said.

“As data processing and reporting becomes quicker, it’s important that focus isn’t just on shorter-term spikes but emerging trends that develop over a longer period which are harder to identify,” says Allianz’s Topham.

Long-term value

Motor insurers must consider external factors, he says, such as economic conditions, climate change and new legislation and the impact to the claims experience.

The latest data from the ABI comes one year after the trade body launched its 10-point Roadmap to Tackle Insurance Costs, which sets out practical steps for the industry and government.

Some key areas included tackling vehicle theft, cracking down on insurance fraud and uninsured driving, helping consumers make informed decisions, and reducing the impact of the personal injury discount rate (PIDR).

Yet the cost of cover in 2024 was £622, 15% (£78) higher than 2023. This is against a backdrop of total claims payouts that were 17% higher in 2024 compared to the previous year.

Rhodri Charles, head of UK motor pricing at Admiral says, “Having access to real-time data is crucial. Numerous factors that influence our pricing, such as market fluctuations or the volume of claims, can change quickly. Our pricing department continuously reviews our data, and having the data in real time better informs our decisions to ensure our prices remain fair and competitive.

Admiral extracts as much value as possible from this information. “Frequent data reporting enables us to identify and monitor trends such as car theft, allowing us to respond quickly and appropriately. We use the findings from these reports to communicate emerging trends with our relevant teams and take timely action for customers based on this information,” says Charles.

In the future, more of this process may be AI-automated as real-time data informs live pricing models. Insurers are recognising the opportunity to enhance existing roles and deploy their resources more effectively.

Human resources

Investing in people, as well as AI and ML tools is a common theme among motor insurers, especially as the role of the underwriter is adapting to the presence of more data scientists and actuaries.

At Axa, underwriters can now utilise the expertise of data scientists to provide advanced analytics and predictive modelling, enabling them to assess risks more accurately and tailor policies to individual customer needs.

While machine learning models and data scientists are crucial for building sophisticated pricing systems, underwriters provide the contextual understanding necessary for markets where data is incomplete or volatile.

Ian Hughes, Consumer Intelligence

“We see significant value in the experience and judgement of underwriters, and we are committed to integrating their insights with data to enhance our processes,” says Axa’s Farrar.

Motor underwriters are evolving into more data-savvy roles, working alongside actuaries and data scientists to bring human judgement to increasingly complex pricing decisions.

Ian Hughes, CEO at Consumer Intelligence, which monitors trends in the insurance market, says: “While machine learning models and data scientists are crucial for building sophisticated pricing systems, underwriters provide the contextual understanding necessary for markets where data is incomplete or volatile.”

He gives the example of the pricing of new electric vehicles, which often involves balancing limited historical data with external trends, such as supply chain issues or evolving repair technologies.

“We believe that advanced tools empower underwriters with the insights needed to make these judgements quickly and confidently,” he says.

For Hughes, in this new dynamic, underwriters are regaining prominence as their expertise complements and enhances the technical capabilities of data-driven pricing teams.

Allianz’s Topham agrees. He sees underwriting as “an evolving profession”, with data and technology enhancing the ability to manage portfolios. What has not been lost in this rush to harness data, he says, is the importance of underwriters building relationships.

“Underwriting continues to be an important part of our industry, and underwriters of the future will need to embrace the technological advancement,” Topham says, “but also continue to have that critical thinking – curiosity and imagination to solve problems”.

Maintaining a competitive edge in motor insurance is on track to be a joint human and AI endeavour.

Sponsored content