← back to rankingTRV · The Travelers Companies, Inc.
Insurance - Property & Casualty · mkt cap $62.1B · calls: Q1 FY2026 vs Q4 FY2025
37.0 conviction · conf-adj 35
conf 4/10 partial
enthusiasm:21.0 · trend:-5 · quantifies:5 · impact:0 · under_radar:0 · credibility:5 · business_impact:8 · disruption:0 · commitment:0 · confirmation:3
Enthusiasm latest 7 / prev 9 (falling)
Travelers' AI thesis is that scale, proprietary data, and underwriting/claims domain expertise let it deploy AI across underwriting, claims, cyber risk control, software development, and distribution workflows. The previous call was much more substantive and quantified, with production GenAI tools, claims automation, headcount/call-center reductions, and workflow productivity metrics; the latest call reiterated the strategy but added fewer new hard numbers. Credibility is relatively high because management tied AI to existing operating processes and concrete efficiency metrics, though direct revenue or margin dollars were not quantified.
GROUNDED NEXT-FY IMPACT vs CONSENSUS
Grounded on actual base — revenue $48.8B · net income $6.3B · net margin 12.9% · diluted EPS 27.43
These are next-fiscal-year annual uplift estimates, not next-quarter numbers.
Aggregate next-FY est. rev uplift: 0.0% · next-FY EPS uplift: % · vs analysts: unclear · priced in: high (model's call-read: medium; verdict above is the hard-data one used for ranking) · confidence: 4/10
| Claim | Figure | Arithmetic | Next-FY Rev % | Next-FY EPS % |
|---|
>$1.5B annual tech/AI investment
other | >$1.5B/yr | $1.5B / $48.828B rev = 3.07%; $1.5B / $6.288B NI = 23.9%. This is SPEND not a benefit — already embedded in current opex/$6.288B net income. Anchored (soft=false) but an input, not an uplift -> pcts null. | | |
>$1.5B AI/tech invested FY2025 (dup)
other | >$1.5B FY25 | Same dollar as row 1, restated for full-year 2025. Investment outflow already in the expense base; no benefit translation possible. | | |
Thousands of AI engineers/data scientists
other · soft | thousands | Headcount color, no $ and no productivity-per-head -> cannot size. | | |
Dozens of GenAI tools in production
other · soft | dozens | Activity/scale indicator, no $ benefit attached -> unanchored. | | |
Millions of transactions automated
productivity · soft | millions | Volume automated, but no cost-per-transaction or labor-saving $ disclosed -> cannot convert to opex saving. | | |
20,000 colleagues use AI tools regularly
engagement · soft | 20,000 employees | Internal adoption metric, no total base or $ productivity value disclosed -> unanchored. | | |
Anthropic partnership: 10,000 staff with AI assistance
productivity · soft | 10,000 staff | Productivity enablement rollout; no $ saving or per-head efficiency disclosed -> cannot size. | | |
>half of claims eligible for straight-through processing
productivity · soft | >50% of claims | % of claims, but no claims-handling/loss-adjustment cost base disclosed anywhere -> no dollar anchor. Directionally lowers LAE but unsizable. | | |
Customers adopt STP ~2/3 of the time
engagement · soft | ~67% take-rate | Take-rate on STP; ~66.7% x >50% eligible implies >33% of all claims via STP. No $ per adopted claim -> unanchored. | | |
Another 15% of claims via advanced digital tools
productivity · soft | 15% of claims | % of claims processed digitally; no cost base disclosed -> unsizable in $. | | |
Claim call-center headcount down a third
cost · soft | -33% headcount | Real cost saving, but neither call-center headcount nor payroll $ is disclosed -> % with no obtainable base. No-invent rule -> null (tax 21% would apply once a base existed). | | |
4 claim call centers -> 2 in 2026
cost · soft | 4 -> 2 centers | 50% facility consolidation; no cost-per-center or total $ disclosed -> unsizable. | | |
Submission intake: hours -> minutes
productivity · soft | hours -> minutes | Cycle-time/capacity gain, no $ and no incremental premium volume quantified -> unanchored. | | |
Renewal underwriting handle time >30% reduction
productivity · soft | >30% AHT reduction | % productivity gain, but underwriting labor cost base not disclosed -> % with no obtainable base; cannot convert to after-tax $ without inventing a denominator. | | |
Assumptions: Incremental net margin = current 12.88% for any revenue uplift; tax rate 21% (after-tax factor 79%) for any cost saving. No phasing — no multi-year dollar target to phase. The single hard dollar figure (>$1.5B/yr tech+AI spend) is an outflow ALREADY embedded in the $6.288B net income base, so it carries no incremental EPS uplift and no new drag. Every benefit claim ('down a third','>30%','>50% of claims') lacks a disclosed cost/revenue base in the inputs, so none can be dollar-anchored without inventing a denominator -> pcts null, soft=true.
Top line: No AI-attributable revenue is quantified. Aggregated adopter-side revenue claims = 0.0% of the $48.828B base. Speed/capacity gains (submission intake hours->minutes, renewal automation) could modestly support premium growth, but management attaches no dollar or volume figure, so any topline lift is unsizable. Travelers sells no AI capacity, so supplier-side is nil (0%).
Bottom line: The AI story is overwhelmingly an efficiency/loss-ratio story: >50% of claims STP-eligible (~67% adopted), >30% cut in renewal underwriting handle time, claim call-center headcount -33%, and 4->2 center consolidation in 2026. These are genuine cost and combined-ratio levers, but NONE carries a disclosed dollar or headcount base, so after-tax savings cannot be sized -> est_eps_uplift_pct null. Direction is favorable but unquantified vs consensus.
Consensus EPS already ramps hard (13.61 FY23 -> 19.04 FY24 -> 25.44 FY25e) and FY25 actuals beat that path (rev $48.83B vs $44.73B consensus = +9.2%; NI $6.288B vs $5.746B = +9.4%; EPS $27.43 vs $25.44 = +7.8%). Strong operating leverage is therefore already embedded in estimates, and the AI efficiency gains plausibly help sustain it. But because management dollar-quantifies none of the benefits, the math cannot show AI pushing results clearly ABOVE consensus (the 'low priced-in / interesting' case) nor confirm it is fully reflected — hence medium. The aggregate adopter-side uplift is null (unanchored), so there is no numeric gap to declare a beat.
MODEL CONSENSUS (impact)
partial
All claims adopter-side; investment is spend already in base; every benefit claim lacks a disclosed dollar base so EPS uplift is null and unsizable.
Conflicts reconciled
- est_rev_uplift_pct: X=0 vs Y=null -> used 0 because no adopter revenue-type claims exist (definitively zero quantified; both converge ~0)
- supplier_rev_uplift_pct: X=0 vs Y=null -> used 0 because the carrier sells no AI capacity (nil, not merely unsizable)
- per-claim soft pcts: X=0 vs Y=null -> used null per schema (unanchored claim -> pcts null)
- $1.5B investment type: X=other vs Y=cost -> used other since it is spend, not a cost-saving
| Field | Opus 4.8 | GPT-5.5 |
|---|
| Rev uplift % | – | 0 |
| EPS uplift % | – | – |
| Priced in | medium | high |
| vs analysts | unclear | unclear |
| Confidence | 3 | 4 |
| Top line | No AI-attributable revenue is quantified. Speed/capacity gains (submission intake hours->minutes, renewal automation) could modestly support premium growth, but management attaches no dollar or volume figure, so adopter-side topline uplift is unsizable (set null, effectively ~0 incremental disclosed). Travelers sells no AI capacity, so supplier-side is nil. | No management claim quantifies incremental AI revenue or premium growth. The hard direct revenue uplift is therefore 0.0% of the $48.828B revenue base; claims such as submission intake and underwriting productivity may support growth but lack a revenue conversion base. |
| Bottom line | The AI story is overwhelmingly an efficiency/loss-ratio story: >50% of claims STP-eligible (~67% adopted), claim call-center headcount -33% with 4->2 center consolidation in 2026, and >30% cut in renewal underwriting handle time. These are genuine cost and combined-ratio levers, but NONE carries a disclosed dollar or headcount base, so after-tax EPS uplift cannot be computed without fabricating denominators — null per the no-invent rule. Note the $1.5B/yr tech+AI spend is already in the $6.288B NI base, so it is a standing investment, not an incremental EPS swing. | The clearest operating claims are claims automation, >33.3% STP-adopted claims, another 15% digitally processed claims, 33.3% call-center headcount reduction, 50% call-center consolidation, and >30% renewal handle-time reduction. None disclose the affected dollar cost base, so EPS uplift cannot be calculated against $6.288B net income without inventing savings. |
| Reasoning | Consensus EPS already ramps hard (13.61 FY23 -> 19.04 FY24 -> 25.44 FY25e) and FY25 actuals beat that path (rev $48.83B vs $44.73B consensus = +9.2%; NI $6.288B vs $5.746B = +9.4%; EPS $27.43 vs $25.44 = +7.8%). Strong operating leverage is therefore already embedded in estimates, and the AI efficiency gains plausibly help sustain it. But because management dollar-quantifies none of the benefits, the math cannot show AI pushing results clearly ABOVE consensus (the 'low priced-in / interesting' case) nor confirm it is fully reflected — hence medium. The aggregate adopter-side uplift is null (unanchored), so there is no numeric gap to declare a beat. | The supplied 2025 consensus was $44.730B revenue and $25.444 EPS; actual 2025 base is $48.828B revenue and $27.43 EPS, which is $4.098B or 9.2% above revenue consensus and $1.986 or 7.8% above EPS consensus. Since quantified AI math adds 0.0% hard revenue uplift and no calculable EPS uplift, there is no numeric evidence here that AI drives estimates ahead of consensus. |
Rows highlighted where the two models disagreed.
QUANTIFICATIONS
Technology investment including AI: more than $1.5 billion annually (Q1 FY2026 / annual, both)
“Our profitability and cash flow support our ability to invest more than $1.5 billion annually in technology, including in our ambitious AI strategy.”
AI and technology investment: more than a billion and a half dollars (full-year 2025, both)
“And during the year, we grew adjusted book value per share by 414% after returning $4.2 billion of excess capital to shareholders and investing more than a billion and a half dollars in cutting-edge AI and other technology initiatives.”
AI technical workforce: thousands (Q4 FY2025 current, both)
“We have thousands of engineers, data scientists, and analysts building AI and other sophisticated technology solutions.”
Generative AI tools in production: dozens (Q4 FY2025 current, both)
“Dozens of scale generative AI tools are already in production.”
Automated transactions: millions (Q4 FY2025 current, bottomline)
“Millions of transactions are now automated.”
Employee AI tool usage: 20,000 colleagues (Q4 FY2025 current, bottomline)
“Within 20,000 of our colleagues use AI tools on a regular basis.”
Anthropic AI assistance rollout: 10,000 engineers, data scientists, analysts, product owners (Q4 FY2025 announced last week, bottomline)
“Last week, we at Anthropic announced a partnership to empower 10,000 of our engineers, data scientists, analysts, product owners with personalized context-aware and integrated AI assistance.”
Claims eligible for straight-through processing: more than half of all claims (Q4 FY2025 current, bottomline)
“In our claim organization, more than half of all claims are now eligible for straight-through processing.”
Customer adoption of straight-through claims processing: about two-thirds of the time (Q4 FY2025 current, bottomline)
“With customers adopting straight-through processing about two-thirds of the time.”
Claims processed with advanced digital tools: 15% of all claims (Q4 FY2025 current, bottomline)
“Another 15% of all claims are processed with advanced digital tools.”
Claim call center headcount: down by a third (Q4 FY2025 current, bottomline)
“As just one example, our claim call center population is down by a third.”
Claim call center consolidation: four claim call centers down to two (2026, bottomline)
“And this year, we'll be consolidating four claim call centers down to two.”
Submission intake automation: from hours to just minutes (Q4 FY2025 current, both)
“Our AI investments to automate submission intake for new business, reduced our time to ingest submissions from hours to just minutes, and we recently extended automation capabilities to renewal workflows.”
Renewal underwriting handle time: more than a 30% reduction (Q4 FY2025 early results, bottomline)
“With early results showing more than a 30% reduction in average handle time.”
PAST (realized)
- Q4 FY2025 | Alan Schnitzer: In extensive testing, we achieved significantly improved engineering output, and meaningful productivity gains.
- Q4 FY2025 | Alan Schnitzer: As just one example, our claim call center population is down by a third.
- Q4 FY2025 | Jeffrey Klenk: Our AI investments to automate submission intake for new business, reduced our time to ingest submissions from hours to just minutes, and we recently extended automation capabilities to renewal workflows.
- Q4 FY2025 | Michael Klein: With early results showing more than a 30% reduction in average handle time.
CURRENT (now)
- Q1 FY2026 | Alan Schnitzer: Our profitability and cash flow support our ability to invest more than $1.5 billion annually in technology, including in our ambitious AI strategy.
- Q1 FY2026 | Greg Toczydlowski: In Select, we have executed some Gen AI that helps us process the business, endorsements, and changes, and just remove the friction and allow it to be much smoother for our independent agent channel.
- Q4 FY2025 | Alan Schnitzer: Dozens of scale generative AI tools are already in production.
- Q4 FY2025 | Alan Schnitzer: Millions of transactions are now automated.
- Q4 FY2025 | Alan Schnitzer: Within 20,000 of our colleagues use AI tools on a regular basis.
- Q4 FY2025 | Greg Toczydlowski: As just one example, we've recently rolled out GenAI agents to efficiently mine both internal and external data sources to better understand and synthesize the risk characteristics and ensure appropriate business classification.
FORWARD (guidance)
- Q1 FY2026 | Alan Schnitzer: Our size gives us the data to power AI and the resources to deploy it, creating a virtuous cycle of better insights, better decisions, and better outcomes.
- Q1 FY2026 | Jeffrey Klenk: The investments we have made in our cyber risk control team for the benefit of our customers—the really good news for them is that as this technology continues to expand and change, we are going to be in an even better position to help them identify and remediate vulnerabilities as they come about.
- Q4 FY2025 | Alan Schnitzer: We expect that this will result in faster and more cost-effective delivery of new capabilities across Travelers.
- Q4 FY2025 | Greg Toczydlowski: We're building and executing a robust portfolio of Gen AI initiatives that will enable enhanced risk assessment and selection ultimately improving loss experience as well as drive gains in productivity and efficiency and improve our industry-leading experience for our agents and brokers.
- Q4 FY2025 | Alan Schnitzer: We're so early in this transformation, which means the benefits more effective underwriting, improved operating leverage, and profitable growth will continue to build.
TRACK RECORD — PROMISE vs DELIVERY
55/100 track record too-early 6 calls reviewed
Travelers makes few hard, dated forward AI targets; most 'AI' content is current-state metrics (STP eligibility, 20k AI-tool users) rather than quantified future guidance. At least one clearly quantified forward AI commitment exists (4-to-2 call-center consolidation in 2026, Anthropic 10k-engineer rollout) but both are too recent to judge, while the operational outcomes cited as delivered were largely reported in-call rather than tracked as prior promises.
Consolidate four claim call centers down to two in 2026, enabled by AI automation/straight-through processing and a generative-AI FNOL voice agent — promised Q4 FY2025
too-early Announced for 2026; the only later call (Q1 FY2026) did not report completion, so not yet judgeable
Anthropic partnership to empower ~10,000 engineers/analysts with AI assistants for productivity gains and faster, cheaper delivery — promised Q4 FY2025
too-early Rollout just announced; no quantified productivity/cost outcome reported in Q1 FY2026
Close 90% of catastrophe claims within 30 days (analytics/automation-driven claims ops) — promised Q4 FY2025
delivered Stated as met in the same FY2025 wrap-up, though reported in-call rather than tracked across later calls
Reduce claim call-center population via automation/STP — promised Q4 FY2025
delivered Reported headcount already down ~a third with >50% of claims STP-eligible, adopted ~two-thirds of the time
PRICED-IN (REFINED)
HIGH (already in)Est. revisions rising · Fwd P/E 21.5 · EV/Sales 1.4x
AI claim maps to Business And International Insurance, Personal Insurance, Bond & Specialty Insurance
Price targets are being raised, with last-month average above last-quarter and last-year averages, and consensus EPS rises sharply across forward fiscal years, even though rating counts remain mostly Hold-heavy. A 21.5x forward P/E is rich for a mature P&C insurer, while EV/Sales near 1.4 is not extreme but still not a bargain signal. AI efficiency would most plausibly flow through underwriting, claims, pricing, and service productivity in Business And International Insurance, Personal Insurance, and Bond & Specialty Insurance. Rising estimates make the AI thesis more priced-in, and paired with the richer forward valuation this points to a high priced-in verdict.
COVERAGE — ENTHUSIASM TRAJECTORY + CATALYSTS
3Q4 FY20243Q1 FY20252Q2 FY20258Q3 FY202510Q4 FY20257Q1 FY2026
AI enthusiasm across 6 calls — trend ↗ rising
AI moved from generic tech investment to explicit, scaled automation and generative AI embedded across claims, underwriting, engineering, and service.
RECENT AI CATALYSTS & NEWS
BUSINESS IMPACT - QUALITATIVE MATERIALITY
7/10 qualitative impact material medium-term · mixed evidence
Where AI matters: underwriting, claims automation, pricing, agent workflows
Travelers is using AI in core insurance workflows, not just generic office productivity: claims straight-through processing, submission intake, renewal workflows, pricing segmentation, call-center automation, and agent friction reduction. The evidence is scaled and operationally relevant, but the company has not tied it to explicit loss-ratio, expense-ratio, revenue, or EPS dollars, so this is material rather than transformational.
Caveats: Benefits may already be embedded in efficiency expectations and valuation; No disclosed dollar savings or AI-attributable EPS uplift; Competitors can deploy similar underwriting and claims automation; AI-related risks could create new insured losses or policy-language uncertainty
AI DISRUPTION / CANNIBALIZATION RISK tailwind · 2/10
AI does not automate away the core product Travelers sells: regulated risk transfer, balance-sheet capacity, claims payment, distribution relationships, and underwriting judgment. The main threat is competitive catch-up or broker/customer bargaining power from better analytics, not structural deflation of insurance premiums as a category.
OPTIONS / MARKET STRUCTURE
option liquidity: good
proxy inputs — dollar-ADV $418M · beta 0.512 · px $291.86
source: proxy (no options chain on FMP)
FMP /stable/ exposes no options-chain endpoint on this key, so ATM IV, bid-ask spread and open interest are unavailable. Liquidity below is a PROXY from dollar-ADV, beta and price level (a stand-in for option depth), not measured option-market data.
CONFIRMATION — INSIDERS · 13F · LANGUAGE
Mixed — insiders selling, institutions adding, management language 5/10 measured.
INSIDERS selling 29 open-market sell(s) vs 0 buy(s) — net distribution
INSTITUTIONS (13F) adding as of 2026-03-31: 169 new / 151 closed positions; 812 increased / 642 reduced; institutional ownership +0.16pp; +15 net 13F holders
MGMT LANGUAGE 5/10 measured AI was barely discussed; they own the investment and resources, but offer no AI-specific outcomes, timelines, or quantified impact.
commit “invest more than $1.5 billion annually in technology, including in our ambitious AI strategy.”
commit “Our size gives us the data to power AI and the resources to deploy it”
hedge “Over time, companies that can leverage scale effectively will have a meaningful edge in consolidating industry premium.”
VERBATIM AI QUOTES
“Our profitability and cash flow support our ability to invest more than $1.5 billion annually in technology, including in our ambitious AI strategy.”
— Alan Schnitzer, Q1 FY2026
“Our size gives us the data to power AI and the resources to deploy it, creating a virtuous cycle of better insights, better decisions, and better outcomes.”
— Alan Schnitzer, Q1 FY2026
“We also recently enhanced our property pricing models, refining catastrophe and non-cat segmentation.”
— Greg Toczydlowski, Q1 FY2026
“Our advanced analytics, market-facing tools, and sales enablement capabilities also played key roles in our success, reflecting the competitive advantages these investments continue to build.”
— Greg Toczydlowski, Q1 FY2026
“In Select, we have executed some Gen AI that helps us process the business, endorsements, and changes, and just remove the friction and allow it to be much smoother for our independent agent channel.”
— Greg Toczydlowski, Q1 FY2026
“We think it can be a great facilitator in helping us be more efficient in our existing distribution channels.”
— Greg Toczydlowski, Q1 FY2026
“It is not just on the negative side—it also has the potential to be on the positive side from an investment in resilience and capability to actually address the threat.”
— Jeffrey Klenk, Q1 FY2026
“The investments we have made in our cyber risk control team for the benefit of our customers—the really good news for them is that as this technology continues to expand and change, we are going to be in an even better position to help them identify and remediate vulnerabilities as they come about.”
— Jeffrey Klenk, Q1 FY2026
“Powered by AI, and not too far off quantum computing, the P&C industry is well-positioned to benefit from AI across the entire value chain.”
— Alan Schnitzer, Q4 FY2025
“This generation of AI can understand and on the complex stakeholder interactions, well-defined processes, data-intensive workflows, and massive amounts of unstructured data that characterize our industry.”
— Alan Schnitzer, Q4 FY2025
“Because AI amplifies existing strength, leaders in the domain are best positioned to use it to drive improvement.”
— Alan Schnitzer, Q4 FY2025
“We have thousands of engineers, data scientists, and analysts building AI and other sophisticated technology solutions.”
— Alan Schnitzer, Q4 FY2025
“Dozens of scale generative AI tools are already in production.”
— Alan Schnitzer, Q4 FY2025
“Millions of transactions are now automated.”
— Alan Schnitzer, Q4 FY2025
“Within 20,000 of our colleagues use AI tools on a regular basis.”
— Alan Schnitzer, Q4 FY2025
“Last week, we at Anthropic announced a partnership to empower 10,000 of our engineers, data scientists, analysts, product owners with personalized context-aware and integrated AI assistance.”
— Alan Schnitzer, Q4 FY2025
“This initiative will enhance and accelerate the development of software, analytics, and predictive models.”
— Alan Schnitzer, Q4 FY2025
“In extensive testing, we achieved significantly improved engineering output, and meaningful productivity gains.”
— Alan Schnitzer, Q4 FY2025
“We expect that this will result in faster and more cost-effective delivery of new capabilities across Travelers.”
— Alan Schnitzer, Q4 FY2025
“In our claim organization, more than half of all claims are now eligible for straight-through processing.”
— Alan Schnitzer, Q4 FY2025
“With customers adopting straight-through processing about two-thirds of the time.”
— Alan Schnitzer, Q4 FY2025
“Another 15% of all claims are processed with advanced digital tools.”
— Alan Schnitzer, Q4 FY2025
“As just one example, our claim call center population is down by a third.”
— Alan Schnitzer, Q4 FY2025
“We're pleased with our progress with GenAI.”
— Greg Toczydlowski, Q4 FY2025
“As just one example, we've recently rolled out GenAI agents to efficiently mine both internal and external data sources to better understand and synthesize the risk characteristics and ensure appropriate business classification.”
— Greg Toczydlowski, Q4 FY2025
“Our AI investments to automate submission intake for new business, reduced our time to ingest submissions from hours to just minutes, and we recently extended automation capabilities to renewal workflows.”
— Jeffrey Klenk, Q4 FY2025
“And in our commercial surety flow business, we've leveraged AI to enhance distribution submission and fulfillment experiences, improving efficiency, and fueling growth.”
— Jeffrey Klenk, Q4 FY2025
“As just one example, we're leveraging artificial intelligence to make our renewal underwriting process more effective and efficient.”
— Michael Klein, Q4 FY2025
“With early results showing more than a 30% reduction in average handle time.”
— Michael Klein, Q4 FY2025
ANALYST QUESTIONS ON AI
Q (Q1 FY2026, Gregory Peters): I am curious how it is affecting the culture of the company. I am thinking about this from two perspectives. First of all, a number of your peers have talked about the potential for headcount reduction. And then at the SBU or line of business level, there are risks, I suppose, of deploying new technology both on growth and margin, and maybe sometimes that might outweigh the benefits.
A: But when you talk about the culture, that is a culture that, fortunately, we have developed and honed over a decade. That is everything from how you pick the right initiatives, how you assess performance along the way, how you measure results, how you prepare an organization to manage change, how you communicate to an organization in the middle of change.
Q (Q1 FY2026, Robert Cox): So I am just curious how The Travelers Companies, Inc. is thinking about underwriting exclusions for AI-related risks and if you are seeing this play out in the market at all?
A: Clearly, we review our policy language all the time when there are new perils or dynamics in the marketplace, and that is evolving right now. We have not had any material changes, but it is something we are watching very closely.
Q (Q1 FY2026, Yaron Kinar): I am trying to think about the impact of AI here, where on the one hand it probably offers an opportunity to increase TAM—you can drive scale and efficiency benefits. But at the same time, it could also mean that we see more of a shift of small commercial to larger brokers with more data and analytics capabilities, maybe greater negotiating power.
A: In Select, we have executed some Gen AI that helps us process the business, endorsements, and changes, and just remove the friction and allow it to be much smoother for our independent agent channel. I do not think it has applicability of just changing distribution channels. We think it can be a great facilitator in helping us be more efficient in our existing distribution channels.
Q (Q1 FY2026, Tracey Banque): I am wondering over time, can commission structures change due to the advancement of AI?
A: It is pretty early, I think, in the evolution of AI and the distribution of insurance to get into that, and it is probably a broader conversation for a different time, different day.
Q (Q1 FY2026, Pablo Zuan): The Travelers Companies, Inc. is one of the largest cyber writers in the U.S., and the question is, how are you thinking about your exposures there and risk management given recent developments with AI?
A: Absolutely, it is an underwriting consideration. We are thinking about artificial intelligence, and with some of the more recent announcements in the last few days about the strength of the LLM models and what that could mean.
Q (Q4 FY2025, Gregory Peters): Just wondering when the structural shift in the expense ratio might materialize. And maybe know, I was looking at you know, the responsible artificial intelligence framework section of your website, maybe you could talk about some of the regulatory and other considerations that might delay some of the expected benefits from your technology spend?
A: We can let it fall to the bottom line if we want through lower expense ratio, we can continue to invest it in other capabilities. Just gives us the flexibility to manage the business. And as I shared in my remarks, the extent that some of these productivity and efficiency benefits are in the claim organization, you know, those come through loss adjustment expense in the loss ratio.
Q (Q4 FY2025, Michael Zaremski): Would you be able to share what you maybe roughly expect your organic headcount growth or shrinkage to be on a percentage basis this year versus, I mean, last year or so?
A: We're not gonna get into projecting headcount beyond that. But what I would say is premium per employee is up, thanks to some productivity and efficiency initiatives. And we expect premium per employee to continue to go up.