AI; helping AXA XL to protect what matters

Authored by Laura Barkatz, Global Lead, Data Strategy and Sandhya Petizon-Louis, Chief Operating Officer, APAC & Europe AXA XL

Artificial intelligence is increasingly becoming an integral part of the way many companies do business. At AXA XL, we are using AI across the insurance value chain – from enabling deeper risk understanding, to informing innovative insurance solutions, to streamlining operations and claims processing.

There is no doubt that AI is transforming the way businesses of all types operate today. In insurance, it is enabling automation and process efficiencies; while also unlocking new insights that enable improved decision-making at every stage of the process, from initial risk assessment and management, right through to risk transfer or claims.

AI is a vital part of our strategy to move from ‘payer to partner’ and protect what matters most for our clients.  At AXA XL we believe that AI can enhance human decision making by providing tools and insights that enable people to do their jobs more effectively and efficiently by focusing on the areas where they add higher value.

To facilitate that, we are developing an ecosystem of AI to help our clients and colleagues in a variety of ways. For example, our Digital Risk Engineer AI tools monitor the health of assets and buildings by capturing and intelligently processing data from connected systems like heating, ventilation and air conditioning. And our Opportunity Pipeline Management tool combines diverse data sources and historic performance metrics to highlight the best opportunities in brokers’ pipelines, so that our colleagues can proactively pursue them.

Innovative uses of AI in insurance

AI can help underwriters to better understand the context of their clients’ risks. A great example of this is the use of AI to enrich geospatial satellite and mapping data to assess wildfire risks faced by our clients. 

Our wildfire risk prevention service, developed in partnership with an environmental intelligence firm, is able to give predictive analyses of wildfire risk using geospatial imagery and more than 20 risk factors, including topography and the proximity of electrical cables to buildings, among others. This enables our risk engineers to more accurately assess the vulnerability of a client’s properties to wildfire risk.

Capturing, integrating and analysing new and additional data sources of this kind is part of a virtuous cycle at AXA XL which continuously enhances the quality of our risk analysis, improving the quality of the advice, guidance and insurance services we offer our clients. 

AI can also improve policy wordings. Recent developments in Large Language Models (LLM) present opportunities for policy wordings and clauses to be created and curated using AI.

AI tools can be used to read and process a wide variety and huge volume or document formats to highlight where wordings should be amended or added to, in line with the latest internal guidance. Over time, as we build up this internal knowledge and use these tools to better manage and use it, we foresee that AI will be used to make real-time document-change recommendations, based on the codified, collective expertise of AXA XL colleagues.

And LLM technology, protected by enterprise security and privacy features, is used to power AXA SecureGPT chat, an internal system comparable to ChatGPT. This easy, compliant and data-protected system, coupled with tailored prompt-engineering training and a curated prompt library, makes colleagues’ working days simpler by enabling them to use AI-assisted email, documentation, code-writing and editing tools, among other things.

Sustainability and ESG 

As businesses work to address the effects of climate change and the need to reduce carbon emissions, as well as other Environmental, Social and Governance (ESG) targets, AI can also play a crucial role in developing sustainable business strategies and practices.

For example, AI can be used to create blueprints of buildings and assess their environmental impact. AXA XL is also using AI alongside geospatial technology to help key clients understand the volume of methane gas their factories produce. This enables our clients to devise their own sustainable policies and practices, as well as to demonstrate their progress towards reducing emissions and meeting sustainability targets to their regulators.

Unlocking the potential of data

To develop successful AI solutions, all businesses need a strong data strategy that is closely linked to the company’s overall business strategy. A deliberate and strategic approach to how data is generated, integrated and managed is a prerequisite of successful AI solutions. Otherwise, as the old saying goes: “garbage in, garbage out.”

In recent years there has been a great proliferation of data. Vast quantities of potentially useful data are collected and made available, including unstructured data like text, speech, video and images. 

To unlock its potential this data must be managed – in a targeted way – in order to power AI-enabled risk decision-making and automation, among other opportunities.

Portfolio modelling is just one example where this increased volume and quality of data is enabling us to improve the way we work.

As our own data ecosystems, as well as those of our clients and vendors, become more sophisticated and diverse, the signals in the data, coupled with AI technology, will enable us to follow the evolution of risks in real time – and to take action.

While the use of data is protected by robust legislation like the EU’s General Data Protection Regulation (GDPR), we also need to be keenly aware of our own ethical stance. It’s vitally important that we use data responsibly. AXA Group has a Responsible AI committee that helps us to focus on best practice and knowledge-sharing in this area.

Partnership is the key

At AXA XL we are risk experts. We know how to connect risk information together and are increasingly using AI to continuously improve our risk analytics and customer service.

In so doing, we also work in close partnership with a range of technology vendors and AI start-ups to accelerate innovation in this space.

Co-creation and co-building AI solutions are key to our digital strategy. The expertise of data and AI  specialists combined with our knowledge of risk and insurance is what enables us to create solutions that not only protect our clients, but also help us provide deeper insights into how to prevent risk events from occurring in the first place.

AI is a vital part of our strategy to move from ‘payer to partner’ and fulfil our purpose to enable human progress by protecting what matters.