Takeaways:
- AXA XL scales AI while retaining human presence
- AXA XL uses AI to engage with commercial insurance clients on risk assessment
- Agentic AI has possibilities, but insurers should keep human roots.
This article is from a longer interview and edited for clarity.
Commercial insurer and reinsurer AXA XL has been keeping a finger on the pulse of AI adoption in the insurance industry. Ashok Krishnan, chief innovation, data and analytics officer at AXA XL, recently spoke at ITC Vegas about the changes AI is creating in underwriting. Digital Insurance spoke with Krishnan about the application of AI to underwriting and commercial insurance, its impact on insurers' workforces, and the new focus on agentic AI.
Where does AXA XL stand in its AI adoption efforts?
We are very focused on how to put human expertise at the center of it. Internally, we use the term 'value amplifier.' We think about how AI can amplify the value of our underwriters or claims handlers and other experts. Another piece is our engagement with startups, vendors, and the wider ecosystem. There is a lot of excitement in terms of AI and technology. At the same time, we are also very conscious that we can't do it alone, which is why we want to work with various partners.
How has AI transformation in the insurance industry also become a workforce transformation?
The way we think about AI and agentic AI and more technology, in the broad sense, is about the technology as a tool to amplify the value of our people, because we are very much a business which is focused on expertise. The expertise is what is differentiating us in terms of giving us the ability to support our clients in handling unique risks, evolving risks, because many of the risks that we are helping our clients to manage are fairly niche. For that, the expertise is becoming a crucial factor, and it's a differentiating factor. We are using AI and other technologies to help with enhancing and amplifying this expertise.
How is AI being applied to underwriting, and what benefits does that yield?
Historically, underwriters have been challenged by the volume of data, especially the volume of data in unstructured formats. They have to spend a lot of time just bringing the data together, data entry, crunching the numbers, stitching it all together for it to make sense. Now we are able, with AI, with machines, to take almost all of this current administrative work and help our underwriters to actually focus on the underwriting part of their job.
That is the engagement with our clients – actual risk assessment, looking at the overall portfolio and making portfolio level decisions. Those are very, very exciting things from an overall underwriting standpoint, and we are quite excited that with the technology now, we are able to address the administrative aspects of underwriting.
What is the next step for AI? Where can further advances be made?
As an industry, P&C and commercial insurance, historically has been lagging behind even other parts of the insurance world. Now, it has an opportunity to catch up. Other parts of the financial industry, such as banking, are significantly ahead of how insurance – and specifically commercial insurance – has embraced technology. Definitely, there's a massive opportunity.
Banking is equally regulated or maybe even more regulated than insurance. If another similar, comparable industry can go through a journey like that, why not commercial insurance? That's a fair challenge. Everybody in this industry should hold a mirror to ourselves.
What would it look like if commercial insurance was operating more digitally?
We still will have the core person-to-person contact. Then the person-to-person contact is focusing more on relationship building. It's focusing more on the risk evaluation, risk assessment. It's focusing more on the portfolio management, and it becomes less about the mechanical and administrative current work. That's where the shift would be a good shift for all of us as an industry.
How was AI presented and discussed at ITC Connect?
There were more than 500 startups and vendors partnering in the event, probably close to 600. I did meet quite a few, and I can't think of one that did not speak about AI. So, clearly AI is on everybody's minds. That's definitely a very good thing. At the same time, I would be interested to see how new innovations, going beyond AI, are coming.
Is AI headed to an agentic focus or other applications?
Definitely agentic AI is something where we see a lot of possibilities on how the future could be quite different. Now I'm quite bullish about it. At the same time, we need to go back to the roots of how we anchor a transformation with people. Because ultimately, it's about, how do we get our people onboarded on a technology? How do we make sure people actually use and leverage the technology for their benefit? That's going to be the differentiator.
