Whereas the remainder of the world falls over itself bolting generative AI into their IT operations, insurance coverage firm Aflac appears to be specializing in sensible makes use of of AI that talk to its core enterprise. The automation of claims processing is a technique the corporate goals to save lots of its employees time and power, together with dashing up funds to recipients.
There are different issues from knowledge administration to privateness that insurers should additionally deal with as consciousness and regulation of these areas escalate. Shelia Anderson, CIO of Aflac US, spoke with InformationWeek in regards to the strategy she is taking in her position and the corporate’s journey with cloud and AI.
What are some key know-how management targets you might have as CIO for Aflac? What are the main considerations that you simply deal with from a strategic perspective?
I’ve been with Aflac now about eight months. I’ve had the chance to evaluate the group and route. We even have an Aflac digital technique that we’re nonetheless following. As I have a look at the group, one of many issues is what I’ll name specializing in the inspiration.
We completely are taking a look at a number of the constructing blocks to know-how to allow that we are able to develop sooner or later. And a few of these issues can be a few of your customary enhancements round IT service administration, implementing ServiceNow, implementing a really sturdy monetary acumen, implementing know-how enterprise administration. These are simply a few the foundational components that we’re specializing in.
Certainly one of our different targets could be very a lot centered on delivering for our prospects and being 100% buyer obsessed. That principally means working in partnership with our prospects and our enterprise companions to ship the worth that they’re in search of of their know-how funding. We’re additionally wanting, in fact, at driving worth by innovation and leveraging of latest capabilities, akin to AI and machine studying. That’s simply considered one of many ways in which we’re driving worth for the enterprise.
Underpinning all of that’s completely taking a look at our expertise technique as nicely. As you possibly can think about, all know-how organizations as we speak are in some type of transformation and in the midst of upskilling, reskilling, and hiring for that future state that they’re shifting in the direction of. These are a number of the key issues that we’re working towards this 12 months.
Are there different elements shaping the insurance coverage know-how panorama, such because the methods shoppers need to work together? Are there considerations inside the trade and the way it features which might be a part of the way you type technique?
Sure. As I discussed earlier than, prospects are on the heart of the whole lot that we do. And naturally, behaviors, buyer behaviors, and people experiences that they’re accustomed to in on a regular basis life. That’s one thing that every one of us in enterprise should try towards as nicely. We’ve got a digital enablement staff that designs most of our buyer experiences and our buyer journey. So, we keep very a lot in tune with our buyer suggestions, within the lens of the client as we’re designing for the long run.
There are numerous new issues now when it comes to knowledge privateness and knowledge as an entire — the way it’s processed, what firms preserve knowledge, and what knowledge firms see. Has this influenced your operations?
As you possibly can think about in insurance coverage on the whole, knowledge is extensively accessible. We’ve got quite a lot of knowledge and data, in fact — that’s one thing that underpins all the work that we do.
We’re taking a look at our knowledge technique. What knowledge do now we have? How will we finest leverage that to serve our prospects? After which, in fact, to make sure that we’re sustaining the integrity of that knowledge and that now we have a safe observe going ahead. We’ve got an entire safety group that completely is concentrated on managing and sustaining the safety of that knowledge in addition to all of our safe coding practices throughout the group.
In order that’s prime of thoughts for us. We’ve got alternatives like many firms do to take a look at enhancing that knowledge, ensuring that we’re retaining solely the information that’s wanted to be retained, and naturally, leveraging that to profit our prospects on the finish of the day.
Has the evolving regulatory area with regards to privateness additionally been prime of thoughts? There are completely different authorized questions based mostly on nations and completely different localities and whether or not or not there’s going to be a US nationwide coverage on that.
Within the insurance coverage trade, the compliance and regulatory side is a giant piece of the issues that now we have as we develop any new functionality. In order that’s all the time prime of thoughts for us guaranteeing that we’re in compliance with the regulatory pointers.
After we introduce new merchandise and fee buildings, all of that inside the insurance coverage trade, now we have quite a lot of filings that now we have to take action we’re all the time in compliance. With respect to the whole lot that we do. After all, on the information aspect, it’s the identical factor, actually taking a look at what’s the minimal quantity of knowledge that’s required to do the job that we have to do. For the information that we do have, we’re guaranteeing the general safety and integrity of that knowledge for our prospects.
Let’s discuss a bit about rising know-how. The shiny object everybody talks about now’s generative AI. What strategy are you taking with that, and the way readily do you have interaction with rising applied sciences?
We’ve got groups inside my group which might be functionally aligned, so I’ve the information and analytics staff that works very carefully with our enterprise account on the analytics aspect. That mixed enterprise and know-how staff works as one staff collectively. We glance to hunt out use instances and alternatives within the market. On the rising know-how aspect, we even have an innovation lab known as Hatch right here at Aflac. We regularly leverage that lab together with sources out of my group to check out a few of these rising capabilities. So sure, now we have some [tech] that we’d say, “Okay, these are ones to look at.”
After which we could have others that we’d transfer into the lab to do some type of testing, proof of idea to see the way it may apply to fixing enterprise alternatives inside the trade particularly for insurance coverage.
Within the trade, with any large functionality like [generative AI], there’s additionally considerations round the usage of it in enterprise, a few of that I might say is the power to generate that language functionality on the fly. It’s all dependent upon the way you’re coaching these fashions within the knowledge that goes into it. And there’s additionally questions round who owns that knowledge, who owns the output.
I believe for ChatGPT, that’s going to be an rising space. The entire focus of generative AI will result in another areas that can want a little bit bit extra focus as firms begin seeking to leverage that within their organizations.
Inside Aflac, not generative AI, however with common AI there are some ways in which we discovered worth throughout our group. We began that journey previous to my being right here at Aflac, so I’m very grateful that that they had already had that imaginative and prescient to begin on the journey of leveraging AI within the enterprise. That began in 2022, after which we carried out it as a mix of AI and machine studying, specializing in our claims processing space, specializing in how will we adjudicate claims and course of claims extra effectively and extra precisely.
If you concentrate on it previous to leveraging the strategy to AI, throughout COVID there was this enormous want to actually proceed transacting enterprise that got here throughout with the pandemic. So in lieu of all of our face-to-face interactions which might be very typical for Aflac, now we have to consider, “How will we digitally allow a few of these buyer experiences in order that we are able to proceed enterprise?”
Our preliminary focus was on the accident, well being, and short-term incapacity traces of enterprise. Our claims processing could be fairly complicated. There are quite a lot of guidelines on insurance coverage, as . Our objective was to simplify that and to try a few of our decrease complexity claims and discover a strategy to extra effectively adjudicate these after which present a extra well timed fee to our policyholder.
In the present day now we have about 46% of our claims which might be truly totally automated. We name it straight-through processing. These are primarily within the wellness space. These, for instance, have been nice instances as a result of they don’t require quite a lot of extra proof. There’s questions and solutions on the declare that we are able to principally settle for with out attestation from our policyholder.
So, a little bit little bit of a extra easy, streamlined course of. If you concentrate on our claims set to start with, we even have, I believe, about 49% of our claims which have payouts of lower than $200. Most of these are what I might name lower-complexity claims. And due to that, now we have an important alternative to automate these claims. That’s the place our focus has been.
Of the sources you employ, do any of them stand out as actually transformative for Aflac or from an insurance coverage sector perspective as game-changing for operations?
In insurance coverage, most individuals will let you know the claims course of could be fairly sophisticated. I believe if you can begin leveraging artificial intelligence with a few of these foundational kinds of conditions, you possibly can study from that and construct upon it, and maybe leverage it in additional complicated conditions sooner or later. It could truly expedite processing sufficient that you would be able to actually shorten your time to adjudicate claims and time to ship finally your fee to your policyholder on the finish of the day.
Proper now, many organizations are reorienting themselves. Are there any know-how sources that in months or years prior have been relied on that at the moment are being shifted away from?
One of many issues that’s a part of our journey, that’s foundational for us is our journey to the cloud. We do have some pivots away from on-prem, hosted service, for instance, and we’re taking a really arduous have a look at our journey to the cloud by shifting our full distributed footprint to the cloud within the subsequent two years. We’re in the midst of that journey. Then we’ve separated out our mainframe journey. These are issues that you’ll generally hear most firms are in the midst of.
What different know-how components are you wanting ahead to? Are there different doubtlessly explorative areas of latest know-how that will be nice in the event that they have been to come back into play?
For us, in fact, it’s the AI journey proper now. We’ve simply hit the tip of the iceberg with respect to this. I believe the extra readability that we get round our knowledge, our knowledge fashions, and our knowledge technique for the long run when now we have extra, cleaner knowledge, we could have limitless alternatives to leverage fashions to higher serve our prospects and to additionally enhance our total operational effectiveness as an organization.
What to Learn Subsequent:
What Does the Arms Race for Generative AI Mean for Security?
Microsoft Brings Generative AI to Low-Code Platform
Q&A: US Patent and Trademark Office’s CIO on Cloud and DevSecOps
Discussion about this post