クリティカルポイント・エピソード40：AIから宇宙飛行士の保険まで：インシュアテック最前線

インシュアテック業界は、常にイノベーションを進め、新しいテクノロジーを取り入れて、新たに出現するリスクに向けた様々な商品を生み出しています。今回のクリティカルポイント・エピソードでは、ミリマンのアクチュアリーSheri ScottとRobert Eatonが、人工知能やモーションデータから疾病検知携帯アプリや宇宙旅行保険まで何もかも、自身の経験を踏まえた最新のトレンドや今後の機会について考察します。

原稿

アナウンサー：本ポッドキャストは、啓発のみを目的として一般的情報をお伝えするものです。特定の個人のいかなる状況を導いたり決定したりする意図はなく、具体的行動を起こす際には、有資格の専門家に相談してください。本ポッドキャストで示す見解は、スピーカー個人のものであり、ミリマンのものではありません。

Rebecca Driskill:皆様ごきげんよう。ミリマンがお届けするクリティカルポイントをご視聴いただき、ありがとうございます。ミリマンのRebecca Driskillがホストを務める今回のクリティカルポイント・エピソードでは、2023年およびそれ以降に向けて想定されるインシュアテックの動向についてお話します。ミリマンは、インシュアテック分野のリーダーでもあり、弊社コンサルタントは、自動運転車から、生命保険、健康保険、長期介護保険のプレディクティブ分析のためのデータ利用まで、広範なソリューションを研究しています。

本日のゲストは、これらの分野の専門家のSheri ScottとRobert Eatonです。

Sheriは、ミリマンのプリンシパル兼コンサルティングアクチュアリーで、新興市場の保険リスクのプライシングの専門家です。彼女は、テック業界のビッグネームと協力し、自動運転車や暗号通貨などの最先端のテクノロジーに対する商品作りを支援してきました。

Robertは、ミリマンのプリンシパル兼コンサルティングアクチュアリーで、生命保険、長期介護、職場の補完的健康保険商品の専門家です。また、米国アクチュアリー会（Society of Actuaries）のインシュアテック「Plug and Play」ポッドキャストのホストも務めています。

お二人とも、本日はありがとうございます。早速ですが、2023年以降に向けて、お二人が見聞きしてきた最も興味深いテクノロジーについてお話を伺いたいと思います。まずはRobertからお願いします。

The insurtech industry is constantly innovating, incorporating new technology and creating products for emerging risks. In this episode of Critical Point, Milliman actuaries Sheri Scott and Robert Eaton reflect on recent trends they’ve seen and opportunities ahead—everything from artificial intelligence and motion data to disease-detecting phone apps and coverage for space travel.

Transcript

Announcer: This podcast is intended solely for educational purposes and presents information of a general nature. It is not intended to guide or determine any specific individual situation, and persons should consult qualified professionals before taking specific action. The views expressed in this podcast are those of the speakers and not those of Milliman.

Rebecca Driskill: Hello and welcome to Critical Point, brought to you by Milliman. I'm Rebecca Driskill, and I'll be your host today. In this episode of Critical Point, we're going to be talking about insurtech trends we expect to see going into 2023 and beyond. Milliman is an insurtech leader, and our experts have explored solutions ranging from coverage for self-driving cars to the use of data for predictive analytics across life, health, and long-term care insurance.

I'm joined today by two of those experts, Sheri Scott and Robert Eaton.

Sheri is a principal and consulting actuary here at Milliman and an expert on pricing insurance risk for emerging markets. She's worked with some of the biggest names in the tech industry, helping create products for cutting-edge technology like autonomous vehicles and cryptocurrency.

Robert is a principal and consulting actuary at Milliman, with a focus on life insurance, long-term care, and worksite supplemental health products. He's also a host of the Society of Actuaries’ insurtech “Plug and Play” podcast series.

Thank you, Sheri and Robert for joining us. Let's jump right in. I'm really curious to hear what some of the most interesting technology you're seeing is as we head into 2023 and beyond. Robert, can we start with you?

Coming in 2023: More customer engagement, more forms of data

Robert Eaton: Sure thing, Rebecca, and thanks so much for having us on the podcast. One of the trends that I observe in life, health, and long-term care is really the emphasis on customer engagement that a lot of insurtech companies are bringing to the market. This customer engagement we see both in policy initiation, with applications in underwriting, all of the way through policy administration, collecting premiums, and handling benefit changes, all the way to paying claims. And that's a real change, I think, in what we've seen in the past. I think that it's certainly one of the focus areas in the life and health space.

Sheri Scott: And I would say that we've seen that in the property and casualty space as well. In addition to that, it's kind of evolved from providing these customer-facing technology solutions and integrated solutions to delight the customer to actually using data in new ways to prefill all these customer risk-measurement tools that they're using to price the insurance and collect the data they need to price the insurance and issue the policy.

And some examples of new forms of data that's starting to be used, and I think is going to continue very quickly into 2023, are visual images. So in the past, we would use data that was more like flat file data. A yes or a no answer, or the age of your roof, or the age of the individual being insured. And now, it's moved more to visual pictures of the house and the property, and whether there is brush around the house or surrounding the roof, or actually even overhanging the roof. And visual images to measure whether a crash actually did occur on a vehicle.

And in addition to visual imagery, we're seeing motion data that's being used as well. Motion data is very large data that we're talking about now. It takes up a lot of space. But because of all the advancements in technology and the storage space and processing power that we can now use today, we can now process this motion data. Collect it, store it, process it, and build models with it. Examples of some pretty cool motion data we've been working with are driving behavior data that collects data off the vehicle directly, or off your cell phone, in terms of how the vehicle is being driven. We're also working with clients to develop pilot data, how planes are being flown, and whether they're being maintained adequately, and all kinds of new kinds of data forms and formats that are emerging.

Robert Eaton: You know, it's interesting you mention that, Sheri. We see an analogue of that, too, in the life and health space. Sort of an overall increasing in the richness of the data that's available to us. Oftentimes, that's allowed and provided directly from the customers themselves. They agree to allow companies to collect more data on them. Other times, insurers may try to observe through publicly available sources things about their policyholders and their customers in order to better serve them. So definitely see that trend as well.

Designing products for emerging risks

Rebecca Driskill: That's really interesting. So we're seeing new forms of data modeling, different ways of applying platforms and technologies. What about trends in business lines? Are there any sort of emerging risks, emerging areas where we're seeing, I don't know, new technology when it comes to insurance?

Robert Eaton: So when I think of the overarching trends in the insurtech space and in insurance generally, what I've seen is a move from a back-office focus, and now we're kind of moving up to the front office. Maybe it's as credibility of these solutions has improved, with insurance companies as the clients and as the stakeholders. Examples may have been, really early on, fantastic—cloud-based technologies that allowed for a lot of efficiencies in back-office operations. Where I think that's now sort of table stakes. Any company that gets into the business now has to have that kind of tech sitting behind their insurance operation. Now we're talking about transforming the way that the customers perceive insurance companies and insurance more generally, their relationship with insurance companies. So that kind of back-office to front-office transition, I think insurtechs are really riding that wave in the life and health space.

Sheri Scott: Yeah, and in the P&C space we've seen that as well, where they've moved from the back office to more the front office to start not only engaging with the customer differently, but allowing the customer to have a much easier way of purchasing the insurance and identifying what they need coverage for with data prefill and these new forms of data sources that are now becoming available today. And I would go so far as to say you need to actually have these new forms of data sources and these new methods of measuring the risk, because there are emerging risks today.

These include emerging risks that we just haven't had to price before today, like autonomous vehicles. Those could be your car that has self-driving mode features in it. While it's driving in self-driving mode, what's the risk? It's not really how many accidents I've had, or what my driving history has been, or my motor vehicle record as it's the vehicle. And it's the operating system of the vehicle and the decisions the vehicle’s making that's really creating the risk at that point. We're also seeing emerging risks and creating risk transfer mechanisms and insurance products for autonomous trucks. They're fully autonomous. They're on the road today. You may not have seen them, you may have seen them. And what do those cost? What are the risks associated with those?

There's also virtual assets. So virtual assets, some people call it cryptocurrency. Cryptocurrency, if you're not using it for speculation, if you're using it to actually create efficiency so you don’t have to walk around with paper money—and companies are trying to use it in that manner—that's an emerging risk. Someone needs to insure those assets. Someone needs to insure the rigs on which all those assets are maintained. So we're creating cryptocurrency and crypto operations insurance products.

And some other examples that I think are going into 2023 are to cover not only planes, but now space travel. Individuals are traveling to space now. So we're coming up with insurance products for that. So it's some pretty cool stuff. And really, the trick is to think about what customers today, what their risks are and what they need insurance for and helping them design what that is.

Robert Eaton: That's so awesome. Can I come work for you? That sounds great.

Sheri Scott: That's what we do every day.

Robert Eaton: Those are awesome examples.

Applying emerging technologies to insurance

Sheri Scott: I really do think that tomorrow—tomorrow being in the future—my cell phone carrier or cell phone, Apple, will be able to—or Google—depending on what kind of phone you have, Android phone, Apple phone. They have this breadth of information about you, and you can create all kinds of consumer-friendly insurance solutions, and other kinds of customer-friendly tools with this.

Robert Eaton: There was a study done by a group in Atlanta, and they are indeed using kind of how you look at your phone in order to detect Alzheimer's from a really early age. Of course, the implication of that for someone like a long-term care insurer is enormous. And you can imagine integrating that sort of technology into a new long-term care insurance policy that’s sold today that's going to be in force for 80 years. And to be able to say, "Hey, we actually think that there's a chance you may need to see your doctor, because there's indications that you may suffer from Alzheimer's in the future.” And what that would do to the profile of a long-term care policy risk is really transformative. So yeah, I think there's a lot of cool opportunities.

Sheri Scott: Yeah. There's some revolutionary advancements that I'm seeing in terms of what's available to be used, to not only use for insurance but to use for the consumer, to help them better understand what their health condition is, to help them understand what the risks are and how they can react to that and plan for it.

So for example, driving behavior scores, they're collected on almost every individual in this country that has a cell phone. And so why not tell you, "Oh, this is very risky behavior." I have a new teenager in the house that's going to start driving next month. You couldn’t do this 10 years ago, but today I can—I can actually monitor her driving, not for purposes of monitoring it to know where she's going or that she's following the rules, but really to give her a tool to say, "Hey, you did a great job today. Look how well you drove. Look how you handled this situation. Here are some tools to help measure you against other peers your age, or someone who is advanced or someone that's been driving for many years. Here are some tools so you can help learn and grow from that." So there's some really cool models and opportunities out there just to really better society, frankly.

 

Rebecca Driskill: Do you do that with her? Now I'm curious.

Sheri Scott: She will be getting a driving model on her phone next week, yes.

Rebecca Driskill: That's exciting.

Sheri Scott: Yes. Yeah, there are companies, like Life360 has been doing that for about 10 years now. And there's opportunities for doing that, it's just becoming more widely available to consumers now. Like for example, pilots today, once they hit a particular age, it's actually almost impossible to get insurance for them if they own their own plane, even if they fly it recreationally. And that used to be the case for vehicles as well. So many, many years ago, once you hit a certain age, it was very difficult to get insurance. But nowadays, what they do is they do additional tests, and they just test you more frequently and make sure you've got eyeglasses or whatever you need as you age to be able to drive safely. Same thing with pilots. Once you hit a certain age, it's really unfortunate that you just can't fly your plane anymore. Why not give them scores and information about how they drive their plane and about their body and about their risk characteristics? And say, "Hey, if you do this, then we will be able to insure you."

Untapped opportunities: Blockchain and AI

Rebecca Driskill: Sounds like it's a lot more granular than it ever was in the past, in a lot of ways. Well, so all of these sort of amazing and mind-blowing trends that we might see, or that we are seeing over the next few years, is there anything that you both think insurance companies and tech firms could be doing or should be doing, but aren't trying? Robert, do you want to start?

Robert Eaton: Yeah, so maybe they're starting to try in the life and health space, and more so in the health space and it's increasingly true in the life space at insurtech companies. Previously vendors and kind of the business partners to insurers are starting to want to integrate and become an insurer themselves. I think it's relatively new. The prospect for doing this for a life insurance company can be daunting. And certainly it's a good opportunity for actuarial consultants, like we are. But at the same time, there's a whole lot of things about a traditional insurance company or an insurance company in general that maybe are not as nimble as insurtech companies typically like to operate. So that's going to be a tricky one. But I do see companies kind of going that way in order to create some efficiencies in their business vertical.

The other thing that comes to mind, Rebecca, is I wonder where smart contracts and other blockchain-based technology will find their way into life and health insurance, and even microinsurance. I think in microinsurance, there may be a more straightforward application, as a lot of those events are things that are confirmable maybe from internet sources, or maybe from certain, I think they call them oracles, which sort of determine if an event has happened or not. In the life and health space, it may be a little farther off until we get there. But I can see both among B2B and B2C applications of blockchain and smart contracts in life and health insurance. Hopefully we'll start to see more of that.

Sheri Scott: And we've seen a lot of that in the P&C space as well over the last year and a half to two years already. And I think the goal that we're going to see in terms of companies trying to achieve in 2023 is integrating these efficiencies across the whole value chain, like Robert's talking about. For example, we are already working with a client that's using blockchain to stand up a new insurance company, just to create those efficiencies. And the whole value chain. They're going to serve not only the insurance company back-office operations, but they're going to have the front-end technology that you really need to have to operate in today's world successfully.

But I do also see a trend towards use of more AI. It's becoming much more prevalent across the industry, not only for the insurtech companies, but for the insurance companies as well. One of the drawbacks in the past to using AI has been the lack of volume of data for any one insurance company, or any one MGA if you want to call it, or insurtech company. And so I've seen a lot of new third-party data sources being integrated into these models. And once they have these new model data sources, they now have sufficient credible data to be able to create these models. So Milliman's been working with a lot of companies to aggregate this data and create some opportunities to help our clients create these models and inform themselves with these models. But also, these third-party data sources that we're adding are really enriching the opportunity to create this AI.

For example, weather data. Traffic data. All kinds of additional data that help inform why there was a crash in automobile insurance, or why there was a loss on a house, and whether that claim that just came in for the roof damage or the fence damage really was as a result of the tornado, or if it actually was there before. And so there's all kinds of additional data sources that we're working with our clients to aggregate together so that we can create these opportunities into 2023.

Advancements as actuaries work with data scientists

Rebecca Driskill: So that’s really interesting. I'm curious actually how collaboration is playing a part in terms of advancement when it comes to insurtech.

Sheri Scott: What really has created this evolution and this boom, in my opinion, into the insurtech space, has been the data scientists and data technology companies working hand-in-hand with the actuaries that have traditional rate-making experience. Really, it's them working together that has created this. And that will give the opportunity, like Robert's saying, of being able to rely on traditional actuaries that have done this before, or the actuaries relying upon these new tech companies that have these ideas and are challenging the norm. That's where these new ideas are going to come from, and that's where these advancements are going to come from. And what we specialize here, in our group at Milliman, is to help companies create these insurance products, both the models that inform the insurance products, but the whole product itself as well.

Because the important thing is the consumer, in the end. And the consumer, in my opinion, has somewhat been forgotten about over the last 20 years in the insurance industry. And it's really more now we're putting a lot of attention on consumer-facing solutions and consumer-delightful experiences. And doing that in a profitable manner. Up until now, in the P&C space, I do believe the insurtechs have accomplished the delightful customer experience and front-facing technology. They've done it. They're there already, but they didn’t necessarily do it profitably. And so now that the turn is going to be towards, well, now how do we make that profitable and cover the whole value chain and create efficiencies across the whole insurance value chain for the insurance company operations as well. I really think that's where the opportunity is now, in addition to continuing to evolve to meet or even exceed consumer needs. Don't fall behind.

Rebecca Driskill: Sheri and Robert, this has been awesome. A really interesting conversation, so thank you both for joining me. To learn more about our insurtech solutions, visit Milliman.com. You've been listening to Critical Point, presented by Milliman. If you enjoyed this episode, rate us five stars on Apple Podcasts, or share the episode with your colleagues and friends. We'll see you next time.