Hello, everyone, and welcome to the Lemonade Fourth Quarter 2023 Financial Results Call. Thank you for standing by. My name is Daisy, and I'll be coordinating the call today. I’d now like to hand the call over to our host, Yael Wissner-Levy, VP of Communications at Lemonade, to begin. So, Yael, please go ahead.

Good morning, and welcome to Lemonade's fourth quarter 2023 earnings call. My name is Yael Wissner-Levy, and I’m the VP of Communications here at Lemonade. Joining me today to discuss our results are Daniel Schreiber, CEO and Co-Founder; Shai Wininger, Co-CEO and Co-Founder; and Tim Bixby, our Chief Financial Officer. A letter to shareholders covering the company's fourth quarter 2023 financial results is available on our Investor Relations website, investor.lemonade.com. Before we begin, I would like to remind you that management's remarks on this call may contain forward-looking statements within the meaning of the Private Securities Litigation Reform Act of 1995. Actual results may differ materially from those indicated by these forward-looking statements as a result of various important factors, including those discussed in the Risk Factors section of our Form 10-K filed with the SEC on March 3, 2023, our Form 10-Q filed with the SEC on November 3, 2023, and our other filings with the SEC. Any forward-looking statements made on this call represent our views only as of today, and we undertake no obligation to update them. We will be referring to certain non-GAAP financial measures on today's call, such as adjusted EBITDA and adjusted gross profit, which we believe may be important to investors to assess their operating performance. Reconciliations of these non-GAAP financial measures to the most directly comparable GAAP financial measures are included in our letter to shareholders. Our letter to shareholders also includes information about our key performance indicators, including customers' in-force premium, premium per customer, annual dollar retention, gross earned premium, gross loss ratio, gross loss ratio ex-CAT, and net loss ratio, along with a definition of each metric and why each is useful to investors and how we use each to monitor and manage our business. With that, I will turn the call over to Daniel for some opening remarks. Daniel?

Good morning, and thank you for joining us to discuss Lemonade's Q4 results and to offer some perspective on both the outgoing year and the year ahead. As you will have seen, Q4 was an excellent quarter, capping off a year of dramatic progress for Lemonade. Our top line grew 20% to $747 million of in-force premium, while our quarterly loss ratio came in at 77%, down 12 points from Q4 2022 and down 19 points from Q4 2021. Since Q4 of last year, our adjusted gross profit has nearly doubled, while our adjusted EBITDA loss nearly halved. As I say, dramatic progress. Moving from the income statement to the cash flow statement, it's noteworthy that we're ending this quarter with a total of $945 million in cash, cash equivalents, and investments. That is the very same level we reported at the end of the last quarter, and it is up since our report of two quarters ago. While we expect this level to dip somewhat in 2024, we anticipate our total cash in investments to turn positive again in the first half of 2025 and expect it to dip by less than 10% before reaching that point. Underpinning our results was a steady stream of improvements in our ability to match rate to risk and in our operational efficiencies, all mediated by a singular integrated system that improves and is improved by all our customer interactions. In many ways, therefore, 2023 was the year when the plan came together, when the thesis of Lemonade transitioned from being a hypothesis to being more evidence-based. This isn't a mission accomplished moment, not by a long shot, but the progress in 2023 was tangible and material, and it increases our confidence that we're on track not only to turn cash flow positive next year with plenty of cash in the bank, but to build a large, enduring, and profitable business thereafter. We hope our latest results boost your confidence alongside our own. 2023's results are all the more noteworthy for the turbulent times in which they materialized. The last couple of years were some of the toughest for both established insurance companies and up-and-coming tech companies. As dual citizens, we were buffeted by the storms that afflicted both insurance and technology. As we reflect back on this tumultuous period, we find resonance in the famous words of Nietzsche, or Kelly Clarkson, if you prefer, 'What doesn't kill you makes you stronger.' We are quite sure that we are emerging from these shocks the better for having endured them. We believe we are leaner and more focused, stronger and more resilient with better unit economics and fewer competitors than would have been the case had the turbulence never come. As the African proverb says, smooth seas never made a skillful sailor. As we look forward to 2024, there's reason for optimism that the worst of these storms may be behind us. Inflation appears to be receding, costs of capital may have peaked, and rates are finally catching up with risks. If the headwinds indeed become tailwinds in 2024, that will, of course, be good news. That said, piloting with tailwinds comes with its own set of challenges. And to explore these, let me hand over to Shai. Shai?

Thank you, Daniel. In 2023, we intentionally slowed our growth to minimize sales of products in areas where we're underpriced. At nearly 20% annual growth, we outpaced most of the industry, yet for us, it was a slowdown compared to previous years, and if things go to plan in the coming years as well. And as more rate updates come into effect, we'll have more products in more areas where we can accelerate our growth. To be clear, though, our 77 loss ratio this quarter should not be taken as an indication that our work on rate adequacy is done. We still await significant further rate approvals. And so for much of 2024, we will continue to constrain sales of two products with the highest average premium and the largest markets, Home and Car. In other words, we will continue to throttle growth this year too. That said, the tide has definitely turned. With every passing month, we are seeing more and more opportunities for profitable growth across our portfolio, including Home and Car, and we will be relaxing these growth constraints accordingly. This is great news. And it's the reason we're projecting to begin to accelerate our growth this year. Growth may be a virtue in its own right, but for us, it's a necessity. Our business is still subscale, and we need to continue growing. In 2024, we plan to accelerate our growth rate, but beyond being a welcome sign of improving conditions, accelerated growth also presents challenges we want you to be aware of. Specifically, in order to accelerate growth, we are planning to significantly increase our marketing spend in 2024 versus last year, doubling it actually. We believe that we will generate about a 3x return on the additional $55 million planned for marketing this year, but it's important to remember that this ROI won't materialize within the same accounting period. In other words, we will be spending dollars that will positively boost our EBITDA over time, but will be a drag on it in 2024. In the past, that drag translated into a cash flow gap, but our synthetic agent build has largely taken care of that. Thanks to Synthetic agents, most gross dollars spent in 2024 won't impact our cash during the year. Nevertheless, as GAAP accounting doesn't always follow the cash, our increased growth spend will be registered as a $55 million expense in 2024's EBITDA. All else being equal, that could have translated into year-over-year EBITDA deterioration, but we believe that not all else will be equal. We expect our loss ratio to continue to improve in 2024 and that our new generative AI tech, as well as many other technology innovations currently in the pipeline, will allow us to grow faster with minimal impact to our OpEx. We believe these improvements will more than offset our growing marketing expenses during the year. Indeed, growing the top line while improving the bottom line is a tricky balancing act, but we plan to pull it off in 2024. To give a better sense of what 2024 might look like and to share more specifics about the outgoing quarter and year, let me hand it over to Tim. Tim?

Great. Thanks, Shai. I'll review highlights of our Q4 and full-year results and provide our expectations for the first quarter and the full year 2024 for the first time, and then we'll take some questions. It was a strong quarter across the board with excellent loss ratio improvement coupled with rigorous cost control, resulting in strong results exceeding our own expectations. Zooming out as noted in our shareholder letter, our actual IFP results in 2023 came in nearly $50 million better than our initial expectations shared a year ago, while EBITDA came in nearly $70 million better. Zooming back in, in-force premium IFP grew 20% in Q4 as compared to the prior year to $747 million. Customer count increased by 12% to just over $2 million as compared to the prior year. Premium per customer increased 7% versus the prior year to $369, driven in roughly equal parts by rate increases as well as mix shift to higher-priced products. Annual dollar retention, or ADR, was 87%, up 1 percentage point since this time last year. We measure ADR on an annual cohort basis and include the impact of changes in policy value, additional policy purchases, and churn. Gross earned premium in Q4 increased 20%, as compared to the prior year, to $181 million in line with IFP growth. Revenue in Q4 increased 31% from the prior year to $116 million. The growth in revenue is driven by the increase in gross earned premium, a slightly lower rate of seeded premium under our quarter share reinsurance structure, and a near doubling of investment income. Our gross loss ratio was 77% for Q4, as compared to 89% in Q4 2022 and 83% in Q3 2023. The impact of CATs in the fourth quarter was roughly 5 percentage points within the gross loss ratio and was nearly all driven by winter storm activity. Absent this total CAT impact, the underlying gross loss ratio ex-CAT was in line with the prior quarter and roughly 6 percentage points better than the prior year. Prior period development was roughly 1.5 percentage points favorable in the quarter. Notably, prior period development from CATs was about 1.5% unfavorable, while non-CAT prior period development was about 3% favorable. Given the notable ups and downs of the quarterly loss ratio, it's all the more useful to also consider the rolling four-quarter view of loss ratio, or TTM loss ratio, that we included in our shareholder letter to get a feel for the longer-term positive trend for loss ratio that we're experiencing. From a product perspective, loss ratio has improved across the business, with the loss ratios of each of the four products we underwrite—Renters, Home, Car, and Pet—all improving between 9 percentage points and 18 percentage points year-over-year. Gross profit and adjusted gross profit have shown notable improvement over time, driven by continued premium growth coupled with loss ratio and investment income improvements. Q4 gross profit increased by 165% to $34 million versus the prior year, while adjusted gross profit increased by 97% over the same period. Zooming out a bit, quarterly gross profit has more than quadrupled in just two years, while quarterly adjusted gross profit has nearly tripled. If loss ratio improvement continues as expected, this recent trend of adjusted gross profit growing faster than our top line may well continue for some time. Operating expenses, excluding loss and loss adjustment expense, decreased 5% to $90 million in Q4 as compared to the prior year. Other insurance expense grew 18% in Q4 versus the prior year, a bit less than the growth of earned premium, and is primarily in support of our increased investment in rate filing capacity. If I had to summarize the financial performance year-on-year, I would highlight this: top line up 20%, adjusted gross profit up 97%, operating expenses down 5%, and EBITDA loss 44% improved, an impressive combination. We saw a consistent improvement across the other three expense lines—sales and marketing, technology development, and G&A—all coming in lower than the prior year in absolute terms during a year of notable growth. Total sales and marketing expense declined by $3 million or 10%, primarily due to lower personnel costs related to efficiency gains. Total growth spend in the quarter was $13 million, in line with the prior year quarter. In July, we began to use our Synthetic Agents program, which financed about 50% of our Q3 and Q4 growth spend, and we've upped that ratio to 80% since January 1, 2024. As a reminder, you'll see 100% of our growth spend flow through the P&L as always, while the impact of the new growth mechanism is visible on the cash flow statement and balance sheet, and the net total impact is roughly $15 million at year-end 2023. Staying with the Synthetic Agents program a moment longer, we're pleased to share that we've extended our agreement with General Catalyst by another year and added $140 million to this program. The agreement that was due to renew at the end of this year is now set to come up for renewal only at the end of 2025. Back to costs, technology development expenses decreased 10% to $20 million, due primarily to a reduction in personnel-related costs, while G&A expenses decreased 7% as compared to the prior year to $29 million. Personnel expense and headcount control continue to be a high priority. Total headcount is down about 8% as compared to the prior year, at 1,258, while the top line IFP, as noted, grew about 20%. Net loss was $42 million in Q4, or a loss of about $0.61 per share, about 35% better compared to the $64 million loss, or $0.93 per share, we reported in the fourth quarter of 2022. Our adjusted EBITDA loss was $29 million in Q4 as compared to the $52 million adjusted EBITDA loss in the fourth quarter of 2022, about 44% better. Our total cash, cash equivalents, and investments ended the quarter at approximately $945 million, reflecting primarily a use of cash for operations of $119 million since year-end 2022, while total cash equivalents and investments is down only $92 million in that same period. As Daniel noted, total cash and investments actually increased by about $3 million since June 30, 2023. Now with these goals and metrics in mind, I'll outline our specific financial expectations for the first quarter and for the full year of 2024. Please note, our approach to guidance this year is measured as always, but it's realistic and it aligns with our own internal expectations. While we certainly delivered excellent results in 2023, particularly when compared to our original guidance for the year, we don't currently expect that same magnitude of overachievement in 2024. While there is always the aspiration for outperformance, especially this early in the year, we would urge listeners to exercise caution about assuming that our guidance is overly conservative. It is not. For the first quarter of 2024, we expect in-force premium at March 31 of between $789 and $791 million, gross earned premium of between $183 million and $185 million, revenue between $111 million and $113 million, and adjusted EBITDA loss of between $43 million and $41 million. We also expect a stock-based compensation expense of approximately $15 million in the quarter, capital expenditures of approximately $2 million in the quarter, and a weighted average share count of approximately 70 million shares. For the full year of 2024, we expect in-force premium at December 31 of between $938 million and $942 million, gross earned premium between $815 million and $820 million, revenue between $505 million and $510 million, and an adjusted EBITDA loss between $160 million and $155 million. Stock-based compensation expense for the full year, we estimate to be approximately $60 million. Capital expenditures, approximately $10 million, and a weighted average share count for the full year of approximately 71 million shares. And finally, I note that we've published a handy deep dive guide to a couple more detailed accounting topics in conjunction with our shareholder letter. You'll find the supplementary information on our investor relations website as well. It gives some additional detail on our cash flow, our synthetic agents accounting program, and reinsurance. It's worth a read. And with that, I would like to hand things back over to Daniel to answer some questions from our retail investors.

Thanks, Tim. We'll now turn to questions submitted and upvoted by our community of engaged and extremely thoughtful shareholders. And the first one comes from 'paper bag' who asks, with car loss ratios improving, how aggressively will Lemonade expand this product? What percentage of IFP is car now, and what might it be in a year? So at year's end 2023, Car represents about 15% of our total IFP, and we expect it to roughly maintain that share in 2024 and begin to expand significantly in 2025 and beyond. Put differently, Car is going from losing share in 2023 to really carrying its own weight and maintaining its share in 2024 to growing its share thereafter. In terms of how aggressively we plan to expand, let me say the following. First, you're right, of course, as the loss ratio of Car comes down to our target range, we will be seeing opportunities to invest much more aggressively in growing the book. That process has begun, and it will step up in 2024, but it won't really hit its stride until 2025. But as I say, it has begun. In fact, in 2023, Car actually grew significantly in places where profitability was attractive. Now, that wasn't true for our largest markets like California and New Jersey, but outside of those states, our book is relatively small, but it grew at a 50% growth rate in 2023. So where opportunities present, we can grow pretty aggressively. And this foray into growing Car has given us data, experience, confidence, and time to adjust rates in those other places. So we will be increasing growth investment in 2024, and we're expecting Car to account for 10% to 15% of newly written premiums this year. That's still not the breakout we expect Car to deliver in due course when the engine is finally tuned and revved up, but we're definitely gearing up towards that. There are several exciting technology advancements this year that should help with cross-selling and further refinement of our telemetry and the data science behind that; those will be released in 2024 and enable us to accelerate car growth further in 2025. Given that that's the case, we currently project that the share of newly written car premiums next year will probably be double what it is this year. I hope that addresses your question. The next one comes from Darren. Darren asks two related questions, so I'm going to bundle them together. Firstly, he says, in Q1 2023, Lemonade had mentioned that generative AI was meaningfully improving the cost structure or was expected to over the course of 18 months, and now we are about 12 months into those 18 months. Are there any financial benefits that have been realized, any evidence for how generative AI can help? And he also asked if we could elaborate on how much IFP and customers the current headcount could support, particularly if we had a $10 billion or when we have $10 billion of IFP, how much larger would our employee base need to be? So Darren, those are really great questions. They're exactly the kinds of things that preoccupy us and that we focus on. Let me start by giving you kind of the rearview mirror, what's been achieved so far. We've delivered significant operating leverage in recent years. Over the last few years, headcount has grown by 8% compounded annual growth, while IFP has grown by 40% CAGR during the same time. So you can see those dramatic divergences between an 8% headcount expense or payroll expense and the 40% top line. I think that speaks volumes about the financial benefits we've realized and continue to realize. In fact, our payroll expense was down in 2023 year-on-year, even though our business grew at 20%. Again, I think that's a very strong indicator. Giving a little bit more context and color, when we think about the scalability of our employee base, we tend to think about it as split, roughly half our employees are what we think of as variable costs. Those are the customer-facing teams, customer support, claims that you would typically expect to grow more or less in line with customer growth. Twice as many customers will usually mean twice as many customer support queries and twice as many claims. So that would naturally, all else being equal, grow more or less in a linear fashion. And then there's everything else—the engineering, finance, legal, etc.—where you wouldn't have that expectation. The outstanding historical operating leverage I've mentioned was really delivered by an ever-improving, pretty dramatic improvement in automation rate, placing significant downward pressure on the growth of variable headcount, plus a lot of smart expense management and targeted automation initiatives that keep the rest of the organization more or less in a stable size, marketing, and other teams that are using generative AI, coding, and teams using generative AI, etc. Just two months ago, we referenced that some of our brand new generative AI tools were handling 7% of incoming customer service emails. To be clear, that's distinct from the roughly third of in-app interactions that Maya handles unaided. Emails are much harder; they usually come without context. You don't necessarily know which customer you're talking to, which policies they have, unlike the app where all the context is already there. So that's been a harder nut to crack. At any rate, I'm happy to report that just one quarter later, that 7% has pretty much tripled, roughly 20% as of now. So we're seeing a very notable acceleration. You can see the trajectory that's on. We think of what we've done so far as the early innings, the tip of the iceberg, or whatever other metaphor you want. We see no reason why we can't ultimately automate the overwhelming majority of our customer interactions and to deliver them as we do today with at least the level of customer satisfaction as we generate from human interaction. So we're seeing very similar levels of net promoter score and CSAT and other indicators. Shifting to look forward to kind of how far can this go? Well, in terms of our fixed teams, if you like, what I was referring to as engineering and financing and marketing and others, we do think that is massively scalable. We think we've got very strong teams, well-staffed, and conceptually, we could double, triple, or perhaps even 10x our business without seeing any significant growth in those teams. As we choose to launch new markets and new products, we may have to introduce a bit more headcount, but I think even then, what I said is broadly true. We are pretty much staffed to the levels that we see ourselves needing in those departments for the foreseeable future. In terms of the variable, we do expect to have to grow those teams as we grow, but given the massive automation drive, generative AI, and other tools, we think that this divergence will continue. Our top line will grow significantly faster, and that the growth of these teams will be a fraction of the book's growth rate, and that's really where the scalability and financial results come from. So in summary, all of these anecdotes, statistics, or historical successes we've had are indicators of the operating leverage, how much it's been a focus and will continue to be, and how the sustained scalability of our team is a key driver for how we're focusing on our business. Ultimately, it is that that will deliver the EBITDA breakeven point we've been speaking about in 2026, as we continue to grow our denominator, holding the numerator as close to steady as possible is really the key there, and we think we're doing exactly that. Okay. The next question comes from both Darren and paper bag, pretty much the same question they asked about how things are going with Chewy. Talking about our own results in detail is something we're very comfortable doing. We are far more cautious about disclosing partners' results. They have not made these numbers public, so we're also going to be circumspect. I will say that we are thrilled by our nationwide launch and by the collaboration between the companies. It's very much expanded our reach. We remain very bullish about our ability to continue leveraging that platform. I will also say that the results have so far been broadly aligned with the modeling that we and our partner, Chewy, had done. So without getting too specific, this is pretty much in line, broad strokes, with what we had planned. Beyond the quantity, I also share that the quality of the book has been very good, really on par with our own book in terms of premiums, retention, loss rates, bundle rates, etc. So all of this reinforces our aspirations for this partnership going forward. Finally, paper bag, who got a duopoly of questioners today, asks the following: In Q2 2022, 21% of non-car sales were cross-sells or up-sells. In Illinois, with car, it reached 36%. What's the current cross-sell up-sell rate with car and without? Also, at Investor Day, 3.7% of U.S. customers were multiline; what's the current percentage? Okay. Great. Focused questions. Let me try and address them. So I'm happy to report that all the metrics you referenced are up since the numbers you gave. Over the course of 2023, in the past year, about 25% of our non-car sales were cross-sales. In all other products combined, about a quarter of the sales came from existing customers. If I include car, then we get closer to 30% on a nationwide basis, even though car is not available in all that many states. Nationwide, we're at 30%. Today, about 4.5% of our customers have more than one product, and that's up about 20% from the number you quoted from Investor Day. It's important to note that in Illinois, we have continued to see good progress. We talk about Illinois because it's the first state where all of our products were available. So it's a great indicator of where this might go on a nationwide basis. At any rate, Illinois has more than 9% of customers who have been with us for a couple of years; not more than 9%, almost 10% of them have more than one product. That's twice the average we have elsewhere. It will take time for us to grow a multi-line customer across the entire book, but that might indicate where we might go over the next couple of years in places where we have all the products. We do need to allow the books to mature, continue to expand product availability nationwide, but that gives you some indication. In the near term this year, our focus in growth will be more on Pet and Renters than on Car and Homeowners. I mentioned cars percentages in an earlier question. Excuse me. So we'll see customer growth come mainly from Pet and Renters rather than from the other products, and that won't be beneficial to the numbers they're asking about in terms of multi-line customers. But as Car gets to profitability in different states and as we lean into that, as indicated probably later this year and into next year, then we expect that to also have a significant effect on the multi-line customer numbers in 2025 and beyond. And with that, I'll hand the call back to the operator to take some questions from our friends on Wall Street.

Thank you. Our first question today is from Jason Helstein from Oppenheimer. Jason, please proceed. Your line is open.

Thanks. Two questions. Just the first, you talked in the release about using AI with third-party agents. Is this going to be something that will be financially material in 2024 or is it really more of a test and then it becomes more material in 2025 and beyond? And the second question, you talked about the mix shifting away from home. How much is that due to learnings from the newest LTV model as opposed to just it's taking longer to get rate increases in certain states? Thank you.

Hey, Jason. Good morning. I'm not exactly sure what you mean by that first question. We didn't mention anything about AI with third-party agents. What we did say—I'm not sure if this is what you're asking about—but we said that within home, which touches on your second question, we will be testing selling third-party policies in states where we don't currently offer home. Is that what you were asking about?

Yeah. So yeah, that's what I was referring to, sorry.

Okay. So can you just refine what that first question was regarding that?

Yeah. So is that something that could move the needle in 2024, or is that more of a test, and if it works, we will see the impact in 2025?

Got it. The latter, I mean, these things can take off and move quickly, but it is a test. It will be rolling out just in a few states, and it won't be rolling out until next quarter. So I would not expect it to have a significant impact this year. It's certainly not baked into the numbers that we have provided in our guidance. It's a test that could go well. It's the kind of thing that we've done episodically all along. So earthquake insurance has worked this way. Our term life insurance has worked this way. In areas where we haven't yet got rate adequacy or other things, we're going to test having these alternatives that we can offer to our customers in order to capture a lot of the pent-up demands we are seeing. And that ties into your second question about rate adequacy from home. So we do find or have found that it's taken quite a long time to get to rate adequacy in many states. This is not just about LTV 9 coming through that has allowed us to refine as it does with every generation to understand with greater nuance exactly where and which customers and lifetime values. But frankly, the issues with homeowners insurance in states like California have been so broad-based industry-wide, as we reference the largest insurance companies in the nation pulling out of the state. So there have been larger forces at play—secular shifts—which we've been responding to. Our tools of precision and LTV are doing a fabulous job. But if regulators aren't approving rates that are reflective of those AI insights, then we're not going to sell insurance in those markets until they do. And we're just seeing a time lag, so I think we have a very good understanding of which customers we want, how to select for them, and how to price for them. We're waiting for those rates to come online so we can do exactly that.

Thank you.

Thank you. Our next question is from Yaron Kinar from Jefferies. Yaron, please go ahead. Your line is open.

Thank you, and good morning, everybody. My first question, Tim, it was helpful to hear the updates on the loss ratio progression by line. Can you maybe be a little more specific there? I think you gave a range of 9 to 18 point improvement across each of the lines, but can you maybe give us a little more color onto which lines are improving by how much, and if possible, even without catastrophes?

So we haven't been disclosing exact loss ratios every quarter, but wanted to give some color commentary; significant improvement across all the lines. I would say that Car showed the most significant improvement, so it was at the upper end of that range, and obviously no material CAT impact there typically, so not affecting the number at the lower end of the range—Renters, Home, and Pet. Home tends to be the one where the CAT impact is most significant. So good progress across the board, 9% to 18%. We'll continue to give that sort of color commentary where it's helpful, but it's not something we disclose specific numbers every single quarter.

Okay. Thanks. And then I just want to make sure I'm thinking about this strategically correctly, the homeowner's business. So maybe a little less of an appetite to grow there right now while rates are not quite adequate in all states, but you are still committed to growing in home over time on your own balance sheet in the U.S., is that correct?

That is. Hi, Yaron. Good morning. Yes. We are seeing significant progress in our homeowners business. As Tim indicated, rate adequacy, as you know, has been a challenge for the industry. In the meantime, I say in the meantime, but there will always be certain business that we don't want to underwrite ourselves. We'll have our own underwriting appetite. We do have our own underwriting appetite. And even in places where we are very profitable, particularly in those, there's always the risk of getting over-indexed on those regions. Therefore, having an avenue through which, when we reach our appetite limits, we can continue to satisfy customer needs, either by writing our own paper or elsewhere, is something that just makes a lot of solid sense and is widely done throughout the industry. So this is not a substitute for us doing our own; it is an augmentation that we're testing at this point.

Got it. And then one final one, if I may, on AI. So I've seen some press recently about states that may scrutinize the use of AI and insurance a bit more carefully, both on the claims side and the underwriting side, I think even the customer acquisition side, and just looking at trying to avoid any discriminatory practices that could arise there or seemingly discriminatory practices. Can you maybe talk about what you're seeing there, how you avoid these pitfalls, and how much buy-in you're getting from the regulators?

Yeah, with pleasure. It's a great question. We are seeing some of that in the U.S.; Europe is perhaps a step ahead in this regard in passing AI-related regulation. I think it's fair to say that we've been pushing and encouraging and working with regulators on this from the get-go. We've had for some years now an AI fairness officer, an AI ethicist. We have our data science teams trained on these matters. We've got fairly strict protocols. The old adage about great power requiring great responsibility holds true. If we are, as we claim to be, at the cutting edge of bringing AI to insurance, we also have to be thoughtful about all of those risks that come with it as well. We're quite sure that when it's applied responsibly, the benefits dramatically outweigh the risks, but the risks are not to be ignored. So yes, we do work with regulators on these things. We feel quite comfortable that we are not only within regulation but in large measure setting the standards for how this can be done responsibly, how this can be tested on a nationwide basis, on a per model basis, on a per customer basis. So yeah, I think we're feeling pretty comfortable with that and continue to monitor this space. We recently created an InsurTech coalition that is working with regulators to set the tone and help regulators understand these issues. So, yeah, an area of significant focus.

Thank you.

Thank you. This is all the time we have for questions today. So that will conclude today's call. Thank you, everyone, for joining. You may now disconnect your lines and have a lovely day.