Prompted LinesAI guidance for insurance

Strategy · Leaders & finance · ~10 min

Costs and value: what AI costs, and which use cases pay

The gap between a demo and a P&L line is practical detail. This page answers two leadership questions: what does it cost, and which use cases pay? Examples emphasize specialty and large-commercial work; the operating principles apply across lines. Companion to the integration phases and capability ladders.

Tokens are the unit of cost, and agents spend them like a fleet, not a person. Every call pays for context in and text out; workflows make many calls per task, and fan-out multiplies them. Engineer cost with the same discipline as quality.

The figures below are costed for specialty work, where a single account carries far more document handling than a standardized risk. For why that is, and the carrier results behind it, see Specialty & E&S.

1 · Token economics: where the money actually goes

Four drivers set the cost of an agentic workflow, and none of them is the model's sticker price:

Budget anchors. Mid-size enterprise LLM agreements run roughly $250k to $1M+ a year; volume discounts are material but must be verified in procurement (benchmark guide). Insurer IT spend averages ~4.5% of GWP (Datos Insights), and two-thirds of insurance CEOs plan to allocate 10–20% of budget to AI (KPMG CEO Outlook, PDF). Vendor purchases succeed about 67% of the time versus 33% for internal builds (MIT, PDF), supporting a buy-commodity, build-differentiation strategy.

The discipline

Measure cost per completed task and dollars per underwriter hour returned, not price per token. A use case that cannot clear both bars is a demo. The winning comparison is machine cost per submission versus loaded human cost per submission reviewed, and it usually is not close.

The four drivers multiply, which is why sticker price misleads

None of these is the model's price per token. They compound: halving context and routing two steps down a tier does not add up, it multiplies down.

Context per callWhat each step reads. Send what the step needs, not the file room.
Calls per taskA five-stage pipeline makes five-plus; retries and fan-out make more.
Model tier per stepFrontier costs several times mid-tier. Route by step, not by habit.
Rework loopsRegeneration is spend. Evaluation gates stop it reaching a person.

Worked anchor: orchestrated agent teams have been measured at roughly 15x the token cost of a single agent, because fan-out raises calls per task and each agent pays its own context. Measure cost per completed task, not price per token.

2 · The operational realities the demo never shows

ConsiderationWhat it looks like in productionThe practical rule
Latency and broker SLAsMulti-step agents take seconds to minutes; brokers notice turnaround, not your architecturePut speed where the broker sees it (acknowledgment, triage, appetite answer) and depth behind it
Enterprise termsZero data retention and no-training clauses, audit logging, regional processingNo enterprise agreement, no company data. This is the first control in the governance section
Model deprecation and driftVendors retire and upgrade models; behavior shifts silently under a pipeline that passed its evals in MarchRe-run the golden dataset on every version change, and pin versions where the provider allows it
Rate limits and surgeA CAT event is a volume spike; provider rate limits are a hard ceilingCapacity-plan for surge events, and keep a queue-and-degrade path that keeps humans working when the cap is hit
PII and residencyClaimant and policyholder data in prompts and logs; state and partner rules on where it may be processedMinimize and de-identify by default; log retention is a compliance surface, not an IT detail
Vendor viability95% of H1 2026 insurtech funding went to AI startups (funding data), which means consolidation is comingDue diligence on funding and runway; an exit plan for any vendor whose output feeds a regulated decision
Lock-in and portabilityA harness written to one provider's quirks is a migration project later; both core vendors shipped agentic frameworks in 2026 (Guidewire, Duck Creek)Keep context specs, evals, and hooks model-agnostic; they are your portable assets

3 · The most impactful use cases for insurance companies, ranked

Ranked by measured value divided by implementation risk, from the carrier and vendor results behind the integration phases (verbatim source passages for the top entries are on the evidence page).

#Use caseWhy it pays (measured)Where it sits
1Submission intake and triage2–5x underwriting speed and 370k+ submissions a year at AIG; Markel's 113% productivity uplift; 50–97% faster processing and +15% hit ratios at Sixfold customers (evidence)Underwriting; the proven first move
2Document extraction and summarizationLoss runs, SOVs, claims files: routine 50–80% time cuts on document-heavy work; the foundation every other use case reads fromEverywhere; assistive, low scrutiny
3Claims triage, severity and litigation predictionAttorney-involved claims cost ~4.9x more (CLARA data); early triage moves both cycle time and indemnityClaims; decision support with human authority
4Bordereaux processing85–94% processing time savings (Verodat); the unglamorous pain point of program businessProgram/delegated-authority operations
5Fraud detection5x more fraud detected at Tokio Marine (case study, vendor-reported)Claims; scoring with human review
6Knowledge access (RAG over guidelines)Appetite and procedure answers in seconds; multiplies every other use case by keeping context currentEnterprise-wide; internal only
7Actuarial filing research and pricing workbenchFiling research from weeks to hours (Akur8); 13 pricing tools in 13 weeks at Allianz Commercial (hx)Actuarial/pricing; assistive
8Leakage and subrogationAI pre-payment controls prevent 90–95% of detectable leakage (analysis); $15–20B of subrogation goes uncollected annually (industry estimate)Claims finance; Phase 3 material
9Bounded agentic quoting3 days to ~3 minutes at Hiscox London Market; CFC's agentic pilot (evidence). Real, but gated: human authority, governance gates, bounded segments onlyUnderwriting; last, under Phase 3 gates

The same nine, arranged by how much scrutiny they demand

Grouped using the sequencing rule stated below the table and each entry's own "where it sits" column. Read left to right: the highest-ranked moves are also the least governed, and the one that needs the most governance is ranked last. Numbers are the value-over-risk rank from the table.

Start here · assistive

AI drafts, a person decides. Low scrutiny, highest confidence.

1Submission intake and triage 2Document extraction and summarization 6Knowledge access over guidelines 7Actuarial filing research and pricing workbench

Then · decision support

The model scores or recommends; a named human holds authority.

3Claims triage, severity and litigation prediction 4Bordereaux processing 5Fraud detection

Last · bounded automation

Gated work: bounded segments only, behind Phase 3 controls.

8Leakage and subrogation 9Bounded agentic quoting

Ranking answers what to do first; this arrangement answers what has to be in place before you can. A high rank never licenses skipping the governance a stage demands.

Sequencing rule: start assistive (ranks 1–2), move to decision support with human authority (3–5), and reach bounded automation last (9). That is the integration phases's phase logic applied to a single function; the harness section of the practice ladder is the build manual for each step.