Will AI Replace Underwriters? The Numbers Behind the Shift

Insurance underwriting has always been about sizing up risk. You review an application, weigh the data, check the actuarial tables, and make a call — approve, deny, or modify the terms. It is a job built on pattern recognition and judgment, which is exactly why AI is making rapid inroads. Our data shows AI exposure for insurance underwriters at 64% in 2025, up from 52% just two years ago, with automation risk at 62%.

Those are some of the highest numbers in the financial services sector. But before you update your resume, the full picture is more nuanced than the headline suggests. The US insurance industry employs roughly 120,000 underwriters across personal lines, commercial lines, life, and specialty markets, and the role is bifurcating sharply between routine work that is automating and complex work that is becoming more demanding.

Where AI Is Already Doing the Work

The clearest impact is in routine risk assessment. AI systems can now process standard applications — homeowners insurance, auto policies, straightforward commercial lines — faster and more consistently than human underwriters. These systems pull data from dozens of sources simultaneously, run it against historical loss patterns, and generate pricing recommendations in seconds rather than hours. Carriers like Progressive, Lemonade, and Root have built entire personal lines operations around AI-driven underwriting, with human reviewers handling only the cases the algorithms flag as ambiguous.

Predictive modeling has transformed how carriers evaluate risk. Machine learning algorithms can identify correlations in claims data that no human would spot, from the relationship between specific building materials and fire loss frequency to the subtle patterns that predict auto claims. A senior underwriter at a top-ten carrier told us the models now catch risk factors that even experienced professionals miss — and just as importantly, the models surface combinations of factors that traditional rating plans cannot represent. Telematics data, satellite imagery of properties, and even social-media-derived business activity signals are now standard inputs at progressive carriers.

Document processing is another area where automation is well advanced. AI can extract relevant information from applications, financial statements, inspection reports, and medical records, then flag inconsistencies or missing data. What used to take an underwriter thirty minutes of reading and data entry now happens in under a minute. The downstream effect is that underwriters can review 3-5x more accounts per day, but the accounts that reach them are systematically harder than what they used to see.

Portfolio monitoring has also shifted. AI systems continuously scan existing books of business for emerging risks — a manufacturing client that just received an OSHA citation, a commercial property in the path of changing weather patterns, a medical practice facing new malpractice trends. This real-time monitoring was simply impossible at scale before. Mid-term cancellations and non-renewals can now be triggered by signals AI surfaces that no underwriter would have systematically tracked manually.

Catastrophe modeling integration with underwriting decisions has also accelerated. The combination of higher-resolution climate models, parcel-level property data, and AI-driven aggregation analysis allows carriers to write or decline coverage with a clearer view of accumulated risk than was possible a decade ago.

What Keeps Underwriters in the Game

Complex and unusual risks still need human judgment. When a tech startup wants coverage for a novel product, when a manufacturer is expanding into a country with limited loss data, or when a claim history shows an unusual pattern that could mean either bad luck or fraud, experienced underwriters bring something AI cannot replicate: the ability to weigh ambiguous information and make judgment calls that balance risk with business opportunity. The specialty lines market — cyber, transactional, environmental, professional liability — is where most growth in underwriter headcount is happening.

Relationship management is another anchor. Underwriters who work with brokers and agents are not just processing paper — they are building partnerships, negotiating terms, and making exceptions that make business sense. A broker who brings a borderline account needs a human who understands context, not an algorithm that says no. The wholesale brokerage channel and the excess and surplus lines market, in particular, run on relationships that no AI can substitute.

Regulatory navigation matters more than ever. Insurance regulation varies dramatically by state and line of business, and the rules change constantly. Underwriters who understand the regulatory landscape can structure coverage in ways that meet both carrier guidelines and regulatory requirements, something AI systems struggle with given the complexity and constant evolution of insurance law. The recent NAIC model bulletin on AI use in underwriting has added a new layer: underwriters must now be able to explain why an AI-driven decision was made, in language a state regulator will accept.

The theoretical AI exposure sits at 87% — meaning the technology could potentially handle most underwriting tasks. But observed exposure is only 38%, reflecting the gap between what AI can theoretically do and what companies have actually implemented. That gap exists because of regulatory caution, integration challenges, and the genuine need for human oversight in consequential financial decisions.

Reinsurance and treaty underwriting remains nearly entirely human. The volumes are too low for AI to learn effective patterns, the structures are too custom, and the trust relationships between cedants and reinsurers are too high-stakes. Lloyd's syndicates and major reinsurers still write meaningful business through face-to-face negotiation.

The 2028 Outlook

Projections suggest AI exposure will reach roughly 72% by 2028, with automation risk climbing to 68%. The trajectory is clear: routine personal lines underwriting will become almost entirely automated, and even standard commercial lines will see heavy AI involvement. The underwriters who thrive will be those handling complex risks, managing key broker relationships, and overseeing the AI systems that handle everything else.

The growth of parametric insurance, the expansion of cyber coverage, and the maturation of climate risk products are all creating new underwriter specialties where institutional expertise does not yet exist. These are the corners of the industry where talented underwriters can build careers that compound over time rather than commoditize.

A Day Inside a Modernizing Underwriting Desk

A senior commercial underwriter at a regional carrier described her current week to us: of the seventy-five submissions her team received Monday morning, the AI auto-bound twelve straightforward renewals, declined nine for guideline violations, and pushed the remaining fifty-four to humans. She personally handled the dozen most complex accounts — including a contractor with a difficult claims history that the AI flagged as decline but where she identified mitigating factors. She also spent two hours on the phone with a key wholesale broker working a manufacturing risk through three rounds of revised terms before binding. The AI had drafted three different pricing scenarios for that account; she chose elements from each, modified the language, and made the deal happen. None of that workflow existed five years ago.

Career Advice for Underwriters

Specialize in complex risk classes where human judgment remains essential — think emerging technologies, international exposures, or novel coverage structures. Develop your relationship skills with brokers and agents. Learn to work with AI tools rather than competing against them — the underwriter who can evaluate and override an AI recommendation with sound reasoning is far more valuable than one who simply duplicates what the machine already does. Consider the growing field of AI model governance in insurance, where underwriting expertise meets technology oversight.

Pursue Chartered Property Casualty Underwriter (CPCU), Associate in Underwriting (AU), or specialty designations like RPLU for professional liability or ARM for risk management. Designations matter in this industry — they remain a meaningful signal of professional development.

Frequently Asked Questions

Will personal lines underwriting careers disappear? Largely yes for routine policies. New entrants to the field should not aim for entry-level personal lines roles unless the path leads quickly to specialty work, claims, or product management.

Where is hiring strong? Cyber underwriting, environmental liability, complex property, specialty casualty, and the wholesale and surplus lines markets. These specialties are growing faster than the industry can train people for them.

Is the CPCU still worth it? Yes — designations remain a meaningful signal and the curriculum has been updated to include AI and analytics content. Employers underwriting your tuition is the norm, not the exception. The CPCU continues to be associated with promotions and meaningful compensation increases, particularly when paired with relevant work experience.

What about reinsurance and Lloyd's careers? Reinsurance underwriting and the London market remain among the most resilient corners of the industry. The volumes are too low for AI to learn effective patterns, the deals are too custom, and the trust relationships are too high-stakes. For an underwriter willing to make the move, reinsurance offers technically interesting work and strong compensation that AI is unlikely to touch in the foreseeable future.

Should I learn to code? You do not need to write production code, but enough Python or SQL fluency to query data systems and understand what AI models are doing is increasingly expected for senior underwriting and underwriting management roles. The combination of underwriting expertise and basic data fluency is unusually valuable in the current market.

For detailed automation data on this occupation, see the Insurance Underwriters page.

---

_This analysis is AI-assisted, based on data from Anthropic's 2026 labor market report and related research._

Update History

• 2026-03-25: Initial publication with 2025 baseline data.
• 2026-05-13: Expanded with carrier examples, NAIC model bulletin on AI, specialty lines growth, modernizing underwriting desk vignette, and FAQ.

Related: What About Other Jobs?

AI is reshaping many professions:

• Will AI Replace Compliance analysts?
• Will AI Replace Fraud investigators?
• Will AI Replace Doctors?
• Will AI Replace Chefs?

_Explore all 1,016 occupation analyses on our blog._