McKinsey Sizes Insurance GenAI Revenue at $50 Billion to $70 Billion: An Investor Thesis Takes Shape

From cross-referencing seven consulting firm sizing reports against actual carrier revenue disclosures over two quarters, the gap between projected and realized AI revenue is narrowing faster in distribution than in underwriting. McKinsey’s February 2026 report, “AI in Insurance: Understanding the Implications for Investors,” puts a number on the opportunity: $50 billion to $70 billion in potential insurance revenue that generative AI could unlock. That figure, and the investor-grade framework behind it, shifts the conversation in a meaningful way. Most industry coverage of insurance AI focuses on expense reduction. McKinsey is sizing the top line.

The distinction matters for actuarial work. Expense savings flow through combined ratios in predictable ways. Revenue creation changes growth assumptions, premium volume projections, and the competitive dynamics that drive rate adequacy. When a consulting firm tells investors that AI could generate tens of billions in new insurance revenue, the actuarial profession needs to understand what that estimate contains, how it compares to alternative frameworks, and whether the early data supports it.

McKinsey Sizes the Revenue Opportunity at $50 Billion to $70 Billion

The February 2026 report, authored by Christian Irlbeck, Grier Tumas Dienstag, Leda Zaharieva, and Matthew Scally of McKinsey’s Insurance Practice, targets institutional investors rather than insurance operators. That audience shapes the framing. Where McKinsey’s typical insurance publications address chief underwriting officers and claims leaders, this report addresses portfolio managers and PE partners evaluating where to deploy capital in and around the insurance value chain.

McKinsey’s headline finding: generative AI could unlock $50 billion to $70 billion in insurance industry revenue, with the highest impact concentrated in marketing and sales, customer operations, and software engineering. The report identifies “insurance runs on data yet leans on manual effort” as the structural inefficiency creating the opportunity, and frames AI as the force capable of converting that data intensity into revenue at scale rather than merely reducing the cost of processing it.

The private equity investment context is essential. US insurance-focused PE expanded at a 26% annual rate between 2022 and 2025. European invested capital declined 18% annually over the same period. Investors who prioritized operational value creation achieved IRRs two to three percentage points above peers. The report positions AI as the operational lever that can sustain those returns as deal multiples compress and organic growth slows in mature insurance markets.

Four Subsectors, Four Value-Creation Mechanics

McKinsey breaks the $50 billion to $70 billion estimate across four insurance subsectors, each with distinct AI adoption economics and competitive dynamics.

Brokers account for roughly 70% of PE transaction volume in insurance, though deal activity declined approximately 20% year over year in 2025 as consolidation matures. AI positions brokers to automate submission processing, match carrier appetite algorithmically, and deploy digital assistants for renewals and cross-selling. McKinsey projects that agentic AI will handle renewals for simple risks with minimal human intervention within the forecast period. The revenue thesis here centers on wallet share expansion: brokers that use AI to serve more accounts per producer capture organic growth that compounds through the book of business.

Managing General Agents (MGAs) represent approximately 5% of PE deal activity but show the steepest growth trajectory. US MGA direct premiums nearly doubled from $47 billion in 2020 to $97 billion in 2024, a roughly 14% compound annual growth rate. AI is compressing quoting timelines from weeks to days in commercial lines and from multiple days to hours in specialty segments. The McKinsey thesis holds that MGAs unable to match this pace risk losing capacity appointments to digital-native competitors that can demonstrate superior submission throughput and loss ratio performance to their carrier partners.

Third-Party Administrators (TPAs) have grown at roughly 15% annually over five years, with strong recurring revenue models and deep access to servicing data. That data access positions TPAs well for AI-driven analytics, but McKinsey identifies a structural tension: prevailing commercial models based on headcount or activity volume may constrain the financial upside of automation. A TPA that uses AI to process claims 40% faster earns 40% less per claim under traditional pricing. The revenue opportunity requires a shift to outcome-based or value-based commercial models, a transition EXL’s Q1 2026 results suggest is already underway.

Software Vendors attracted roughly 20% annual investment growth through mid-2025. McKinsey notes a shift from monolithic AI systems to modular architectures where custom models collaborate using open standards. This maps directly to the multi-agent orchestration pattern that carriers like AIG are already deploying, where specialized AI agents handle discrete tasks (submission ingestion, risk evaluation, pricing benchmarking) and coordinate through an orchestration layer. The vendor revenue thesis depends on whether carriers build these capabilities in-house or purchase them, a question that the Q1 2026 earnings cycle is beginning to answer.

Revenue Creation vs. Expense Savings: Two Frameworks, Different Strategic Implications

The McKinsey $50 billion to $70 billion estimate stands in deliberate contrast to Morgan Stanley’s widely cited $9.3 billion AI savings forecast for P&C insurers. The two projections are not contradictory; they measure different things. McKinsey sizes the top-line revenue opportunity across the entire insurance value chain. Morgan Stanley sizes the bottom-line operating income uplift for P&C carriers specifically. One answers “how much new revenue can AI create?” The other answers “how much can AI reduce expenses?”

The strategic implications diverge sharply. Under the expense savings framework, AI is a margin tool. Carriers invest in automation, reduce headcount and processing costs, and the savings flow through to operating income over a multi-year J-curve. Morgan Stanley projects a net negative operating impact of $2.4 billion in 2026 (implementation costs of $3.0 billion against $600 million in realized savings), improving to a $9.3 billion annual uplift by 2030 through a 200-basis-point expense ratio reduction.

Under the revenue creation framework, AI is a growth tool. Brokers use AI to serve more accounts. MGAs use AI to quote faster and win capacity. Software vendors embed AI to capture higher-margin subscription revenue. The capital allocation decisions follow the framing: expense savings justifies efficiency-focused technology budgets, while revenue creation justifies growth-oriented investment that may temporarily increase, not decrease, expense ratios.

Dimension	McKinsey (Feb 2026)	Morgan Stanley (Jan 2026)
Headline figure	$50B to $70B revenue potential	$9.3B operating income uplift by 2030
Scope	Full value chain (brokers, MGAs, TPAs, vendors)	P&C carriers only
Metric	Revenue creation	Expense reduction
Audience	PE and institutional investors	Equity analysts covering public carriers
Near-term economics	Investment phase, elevated spend	J-curve: -$2.4B in 2026, positive by 2028
Carrier expense ratio impact	May increase short-term (growth investment)	-200 bps by 2030 (from 30.5 to 28.5)

From tracking both frameworks against carrier disclosures, a pattern emerges: the expense savings thesis is more immediately testable (carriers report expense ratios quarterly), while the revenue thesis requires tracking distribution metrics, vendor revenue mix, and premium growth decomposition that standard financial reporting does not always isolate. The challenge for actuarial pricing and reserving teams is that both forces operate simultaneously.

Q1 2026 Revenue Signals: From Thesis to Carrier Filings

Three vendors reported Q1 2026 results that partially validate McKinsey’s revenue thesis with disclosed, auditable figures rather than projections.

EXL Service crossed a threshold in Q1 2026: data and AI-led revenues now represent 60% of total revenue, growing 28% year over year against total company growth of 13.8%. Total revenue reached $570.4 million. That AI revenue share has climbed from 38% in 2020 to 55% in full-year 2025 to 60% in Q1 2026, a trajectory that demonstrates AI transitioning from a supplementary offering to the dominant revenue stream. Insurance remains EXL’s largest segment at $194 million (12.8% growth), and CEO Rohit Kapoor noted clients are progressing “from AI pilots to production.” EXL raised full-year guidance to $2.30 to $2.33 billion.

CCC Intelligent Solutions reported Q1 revenue of $281.3 million (up 12% year over year) with AI-based solutions reaching approximately $120 million in annualized run rate, roughly 10% of total revenue. AI revenue is growing at 3.5 times the total company growth rate and contributed approximately one third of overall year-over-year growth. Adjusted EBITDA reached $120.2 million with a 43% margin, up 300 basis points, demonstrating that AI revenue is not only growing but is margin-accretive. CCC reported 6,500 repair facilities using AI estimating and signed enterprise agreements with Liberty Mutual and Allstate for casualty.

Verisk posted Q1 revenue of $783 million with organic constant currency growth of 4.7%. The AI signal is in the product mix: aerial imagery offerings exceeded 30% revenue growth over two years, digital media forensics (AI-powered anomaly detection) onboarded a sixth top-10 carrier, and the company released seven new client-facing AI modules in Q1 with 25 planned for full-year 2026. CEO Lee Shavel described the company as “commercializing Verisk’s multiyear investments in agentic technologies,” and a global insurer selected Verisk as its strategic co-development partner for a digitally native underwriting entity.

These three vendors represent different segments of the value chain: EXL in outsourced operations and analytics, CCC in claims-tech, and Verisk in data and rating. The common thread is that AI revenue is growing at multiples of base revenue in each case, and the growth is showing up in audited quarterly filings rather than pilot announcements.

Agentic AI Reshapes the Productivity Economics

McKinsey’s April 2026 companion report, “Can Agentic AI Finally Modernize Core Technologies in Insurance?,” provides the productivity mechanics behind the revenue estimate. The report projects 10% to 90% productivity improvements across core system modernization phases, with the widest range reflecting variation in legacy system complexity and agent maturity.

The phase-by-phase breakdown reveals where the gains concentrate. Testing, reconciliation, and defect cycle compression show 15% to 90% improvement potential. Discovery and reverse engineering of legacy systems show 20% to 50%. Cutover and operations yield 10% to 40%. Program management and governance deliver 25% to 50%. The report’s key economic insight is that the marginal cost of reuse declines sharply once core agents are built and governed: the incremental cost of modernizing additional products and systems falls quickly because the same agents, patterns, and context layers can be reused across waves and domains.

This maps to what BCG’s competing framework calls the “modernization factory” model, and the convergence between two of the largest consulting firms on this architectural approach strengthens the thesis. McKinsey notes that agents can now read code written in legacy languages, reverse engineer the logic, and convert it into plain English, and that in many cases this work that would take a subject matter expert months can be accomplished in days.

WTW’s March 2026 survey of 59 P&C insurers provides performance data that links these productivity gains to underwriting results. Sophisticated analytics adopters achieved combined ratios six percentage points lower and premium growth three percentage points higher than slower adopters over 2022 to 2024. Carriers implementing agentic AI reported loss ratio improvements of three to five percentage points, translating to roughly $40 million in annual underwriting profit improvement per $1 billion premium portfolio. Over half of surveyed carriers already use generative AI or LLMs in production, with 60% planning AI-augmented human underwriting by 2028.

What Carrier Earnings Transcripts Reveal About AI Revenue

Q1 2026 earnings calls produced the most detailed AI commentary carriers have yet delivered to investors. The disclosures reveal how operators are positioning AI as a revenue enabler, not only an expense reducer, and the specificity is increasing quarter over quarter.

AIG provided the richest detail. In Lexington’s middle market property business, AIG Assist delivered a 30% increase in quoting submissions, a 55% reduction in time to quote, and approximately a 40% increase in binding. CEO Peter Zaffino described AIG’s progression from single-task AI to multi-agent systems through partnerships with Palantir and Anthropic. AI agents improved from operating autonomously for less than an hour with earlier models to as long as 30 hours with advanced models. On claims, Claude achieved 88% alignment with human adjusters on a 100-claim sample without claim-specific tuning, flagging timeline inconsistencies, geolocation mismatches, and document tampering signals. Zaffino stated that AIG “started our AI journey at the core of our business in underwriting, where we felt the impact will be most profound,” and reiterated a target of over 20% operating EPS compound growth through 2027.

Chubb’s Evan Greenberg provided strategic framing rather than granular metrics, but the direction was unambiguous. He described agentic AI as delivering “enterprise solutions” that “will only accelerate, improve, lower cost, make it easier” and projected AI-driven growth in small commercial and E&S segments as “multiple times bigger” over five years. On intermediation costs, Greenberg observed that in an age of AI, “one of the hallmarks of that is that it ought to ultimately bring down cost.” He disclosed spending significantly more personal time on technology than two years ago, noting that without firsthand knowledge of AI capabilities, executives “start to become irrelevant.”

Travelers reported record Q1 core income of $1.7 billion and disclosed a $1.5 billion annual technology budget with Anthropic-powered assistants deployed to nearly 10,000 engineers and data scientists. Over 30,000 employees have access to frontier models through the internal TravAI platform. The AI narrative at Travelers is infrastructure-first: building the foundation that enables AI-driven revenue and efficiency rather than highlighting individual use case metrics.

Progressive reported personal auto market share reaching 18.6%, with CEO Tricia Griffith crediting “strategic investments in technology and media” for enabling significant share capture while maintaining profitability. Progressive’s AI story is less declarative than AIG’s or Chubb’s because its analytics culture predates the generative AI era; the competitive advantage is embedded in pricing segmentation and selection algorithms that have compounded over decades.

The Re-Rating Catalyst: How Markets Are Pricing Insurance AI

Three institutional research notes published in Q1 2026 frame how capital markets are beginning to differentiate carriers based on AI positioning, creating the conditions for what McKinsey’s report implies but does not state explicitly: a potential valuation re-rating for carriers that demonstrate AI-linked revenue growth.

Goldman Sachs in March 2026 upgraded AIG to Buy and downgraded Allstate to Neutral, arguing that commercial insurers are better positioned than personal lines carriers for AI disruption. The thesis: complex, multinational, and large corporate risks require judgment that reduces disintermediation likelihood, while AI amplifies underwriting throughput for those complex accounts. Goldman identified AIG and Chubb as best positioned among carriers, and Aon and Ryan Specialty among brokers.

Bank of America quantified the disintermediation risk more starkly: $15 billion or more in insurance commissions classified as “low complexity” are at risk from AI, with 20,000 to 30,000 independent agents whose work could be effectively performed by LLM-based digital agents. Specific commission pools cited included Progressive at over $6 billion, Travelers at roughly $3.35 billion, and Hartford at approximately $1.25 billion. Insurance distributor stocks had fallen 24% from peak valuations, returning to pre-pandemic levels, and BofA projected organic revenue growth could slip from 3% to 7% down to 1% to 5% under AI pressure.

The venture capital data tells the complementary story. According to Gallagher Re’s Q1 2026 report, AI-labeled InsurTech startups captured 95.2% of all sector funding: $1.55 billion across 68 deals out of $1.63 billion total. The average deal size reached $25.8 million. Corgi Insurance closed a $160 million Series B at a $1.3 billion valuation, roughly doubling in four months. AI-native insurers and B2B SaaS infrastructure companies command 15x to 30x revenue multiples, while capital-intensive carriers trade at 1x to 3x book value or premium multiples. That valuation gap represents the market’s current bet on where AI-driven revenue will accrete.

The Convergence of Consulting Frameworks

McKinsey’s report does not exist in isolation. Deloitte, Oliver Wyman, and McKinsey all identified AI scaling as the top insurance CEO priority for 2026, converging for the first time on this ranking. BCG published its own three-phase AI-first insurer blueprint projecting $35 billion to $60 billion in US cost reductions. Deloitte’s 2026 survey found 76% of insurers have implemented generative AI in at least one function, with life and annuity carriers leading at 82% adoption.

The convergence extends beyond headline projections. McKinsey and BCG both advocate modular, agent-based architectures over monolithic AI deployments. Both identify the discovery-to-cutover modernization loop as the critical bottleneck. Both project that marginal costs decline sharply with agent reuse. The firms diverge on specific numbers and timelines, but the architectural consensus among the four largest strategy consultancies creates a powerful signaling effect for carrier technology procurement decisions and, consequently, for the vendor revenue streams that McKinsey’s $50 billion to $70 billion estimate encompasses.

WTW’s survey adds an important constraint: only 20% of P&C insurers have a well-defined analytics strategy, and just 12% offer regular analytics training. Data quality and IT support barriers each affect 42% of respondents. The gap between consulting-firm projections and operational readiness is real, and only 7% of insurers have reached full AI scale according to Sedgwick’s assessment. The revenue opportunity McKinsey sizes is genuine, but the path to capturing it runs through organizational barriers that most carriers have not yet resolved.

Why This Matters for Actuarial Practice

McKinsey’s revenue-side framing creates several implications that differ from the expense-savings narrative most actuaries have been working with.

Pricing and rate adequacy. If AI enables carriers and MGAs to process submissions faster and bind more business, premium volume growth could outpace traditional projections. Actuaries setting rate indications need to consider whether AI-driven efficiency in distribution changes the competitive dynamics that determine rate adequacy. A carrier that uses AI to quote in hours rather than days may capture better risks through speed advantage, altering selection effects that feed into loss ratio experience.

Reserving and loss development. Carriers deploying AI at scale may show different loss development patterns than the industry averages that populate most reserve analyses. If AI-powered claims triage reduces average claim duration and improves accuracy (CCC reports 98% gross retention and margin expansion alongside AI adoption), the industry development factors that actuaries apply to less-automated carriers may overstate or understate IBNR depending on the direction of AI-driven changes in settlement patterns.

Expense assumptions in ratemaking. The tension between McKinsey’s revenue framework and carriers building AI savings into forward guidance means expense ratio assumptions face competing pressures. Some carriers may invest aggressively in AI (raising expenses short-term) to capture revenue growth, while others pursue the cost-reduction path (lowering expenses gradually). Actuaries making expense loads in pricing need to distinguish which strategy their carrier is pursuing.

Capital modeling and surplus adequacy. Private equity allocating at 26% annual growth rates into insurance AI creates capital formation dynamics that affect surplus adequacy calculations and competitive entry analysis. If AI-native MGAs continue doubling premium every four years (as the $47 billion to $97 billion trajectory suggests), traditional carriers face competitive pressure that standard market cycle models do not capture.

Professional standards. ASOP No. 56 already requires disclosure of modeling assumptions and limitations. As AI transitions from internal tooling to revenue-generating capability, the scope of what constitutes a “model” subject to actuarial standards expands. Actuaries advising on AI-linked revenue projections for investor presentations or regulatory filings will need to apply the same rigor to growth assumptions that they currently apply to reserve and pricing models.

The $50 billion to $70 billion number will draw attention. The analytical value lies in the subsector decomposition, the contrast with expense-focused frameworks, and the early revenue signals from Q1 filings that suggest the thesis is at least directionally supported by auditable data. Whether the full figure materializes depends on organizational execution at a scale the industry has not yet demonstrated, but the capital flowing toward this thesis is already reshaping competitive dynamics across the insurance value chain.