AI in insurance: from claims handling to underwriting
The insurance industry handles one of the largest data volumes of any sector. Policy administration, claims processing, fraud detection, risk assessment and customer communication run on millions of data points every day, and much of that work is still done by hand. We build AI solutions that speed up these processes and improve accuracy, while meeting the strict requirements of the Wft, GDPR and Solvency II.
Practical AI applications in insurance
Insurers process thousands of claims, policy changes and customer requests every day. The processes below are repetitive, data-intensive and prone to error, which is exactly where AI makes a difference without making the human assessor redundant.
Automated claims processing
AI reads claim files, classifies the type of claim (motor, contents, liability, travel) and extracts the relevant details: policy number, date of loss, amount and parties involved. The system carries out an initial assessment against the policy terms and routes complex cases to the right claims handler. Straightforward claims, such as a cracked windscreen with clear cover, can be handled largely automatically with a final human check.
Fraud detection and anomaly analysis
Machine learning models analyse claim patterns for irregularities: unusually high claim frequencies, suspicious timing, inconsistencies between photos and damage descriptions, or combinations of cover that are statistically unlikely. The system flags suspicious files for manual review by the Special Investigations Unit. It does not make decisions about fraud; it points out patterns that call for human attention.
Automating underwriting and acceptance
For applications for new policies, AI assesses the risk profile based on the data submitted, historical claims data and external data sources. The model generates a risk score and a proposed premium, which the underwriter uses as a starting point. For complex risks, such as commercial liability or industrial property, human assessment remains decisive, but the preparatory analysis takes considerably less time.
Policy administration and document processing
Policy documents, clause changes, medical statements and valuation reports are read by AI, classified and linked to the correct file. OCR combined with natural language processing extracts structured data from unstructured documents. This saves policy administrators from manually retyping information from PDFs, scans and email attachments.
Customer communication and service automation
AI chatbots answer frequently asked questions about cover, excesses, claim status and policy changes. The system can take a First Notice of Loss (FNOL) outside office hours and confirm to the customer straight away that the claim has been received. For complex questions, the chatbot escalates to a member of staff, including a summary of the conversation so the customer does not have to repeat their story.
Loss assessment with computer vision
For motor or property damage, computer vision can analyse photos and make an initial estimate of the extent of the damage. The model compares the images supplied with a database of similar claims and gives an indication of repair costs. This speeds up the assessment process and gives the loss adjuster a well-founded starting point. The final assessment remains with the loss adjuster; AI does the groundwork.
Integration with existing insurance software
The Dutch insurance market relies on specialised systems such as ANVA (policy administration for intermediaries and binding authorities), CCS (claims and policy management for non-life insurers), Guidewire (an international platform for claims and policy processing) and SAP FS (financial and actuarial management). In addition, many brokers use AFAS, Exact or sector-specific ERP solutions to run their operations.
AI solutions only become valuable when they fit seamlessly with this existing infrastructure. We build integrations via available APIs, webhooks or, where no API is available, via structured file exchange. The aim is always for the AI layer to enrich existing workflows rather than replace them: a claims handler keeps working in the system they know, but sees an AI-generated analysis, risk score or fraud alert alongside their file.
At binding authority firms, there is an additional layer: data flows through the binding authority channel to the risk carrier. We ensure that AI-enriched data respects existing data standards (SIVI, AFD definitions), so that exchanges with reinsurers and risk carriers continue without disruption.
The technical reality at insurers
Much insurance software has grown organically over time. Legacy systems sometimes have been running for decades and hold decades' worth of policy data in formats never designed for AI processing. That is not an obstacle; it is the reality we work with.
Our approach begins with a data inventory: which data sits where, in what format, and of what quality? Based on that, we design an integration layer that makes the relevant data available to AI models without altering the source systems. This might be an ETL pipeline that runs overnight, a real-time webhook integration or an API gateway acting as an intermediate layer.
The advantage of this approach: the existing system remains untouched, the AI layer can be detached, and if the insurer later moves to a new core system, the AI component only needs to be reconnected, not rebuilt.
Why the insurance sector is particularly suited to AI
Not every sector lends itself equally well to AI automation. Insurance combines a set of factors that make AI implementation effective and measurable.
Extremely data-intensive. Insurers hold decades of historical claims data, customer profiles, actuarial calculations and external data sources. This is precisely the fuel machine learning models need to recognise patterns that human analysts might miss, not because those analysts are less capable, but because the volume is simply too large to review by hand.
A high share of repetitive processes. Policy amendments, claims registration, coverage checks, premium calculations: a considerable part of the daily work follows fixed rules and decision trees. These are processes where AI works not only faster but also more consistently than manual handling. The employee who today keys in a claim form for the hundredth time will tomorrow be the one reviewing AI suggestions and focusing on the exceptional cases.
Regulatory pressure as a catalyst. The Wft, Solvency II and the GDPR set strict requirements for accuracy, transparency and documentation. Well-designed AI systems inherently offer better traceability than manual processes: every decision is reproducible, every step documented, every model versioned. Regulation here is not a brake, but a reason to automate.
Competitive pressure from insurtechs. New entrants such as Lemonade, Alan and national players are building their entire operations around AI-first principles. Established insurers do not need to overhaul their whole organisation, but can use targeted AI implementations to modernise existing processes and stay competitive, with the advantage of their historical data assets and established customer relationships.
Our approach in four steps
An AI implementation at an insurer does not start with technology, but with the process. We follow a proven approach that takes into account the specific requirements of the financial sector.
Process analysis and compliance scan
We map out which processes demand the most capacity, where error margins are highest and which data is available. At the same time, we test feasibility against Wft requirements, GDPR lawful bases and Solvency II reporting obligations. The result is a prioritised list of AI opportunities with a realistic estimate of impact and lead time.
Data inventory and architecture design
We investigate which source systems (ANVA, CCS, Guidewire, SAP FS, ERP) can be integrated, which data is usable as input for AI models, and what the integration layer looks like. The architecture design describes the data flows, model selection, hosting and security measures, aligned with your IT policy and your compliance department's requirements.
Pilot build and validation
We start with one concrete process, typically claims classification, document processing or a customer service chatbot, and build a working pilot. Claims handlers, underwriters or customer service staff test the output for accuracy and usability. We measure error margins, processing times and user acceptance before scaling up.
Rollout, monitoring and handover
Once approved, we roll the solution out to production. We implement monitoring of model performance (drift detection), compliance reporting and an escalation process for edge cases. Your team is trained, the documentation is handed over and a maintenance agreement is put in place. You retain full control and ownership of your data and models.
Test your idea first: a working prototype in 1 day
With OneDayBuild, we turn your idea into something tangible in one day for €1,150, so you can see whether further development is worth the investment. Decide to go ahead with the full build? Then we credit the full cost.
Explore OneDayBuild →Compliance and privacy: Wft, GDPR and Solvency II
AI in the financial sector operates within one of the most strictly regulated environments. That is not a side issue; it is a core part of every project we build for insurers.
Wft compliance and explainability
The Financial Supervision Act requires financial institutions to be able to explain decisions to customers and supervisors. AI models that support underwriting or claims decisions must therefore have explainability built in. We deliberately choose model architectures that provide transparent output explanations: feature importance scores, decision tree visualisations or textual summaries of the factors that led to a particular score. The AFM and DNB expect you to be able to explain how an AI model arrived at its recommendation.
GDPR and processing of health data
In life, healthcare and disability insurance, AI can process special category personal data: medical certificates, health declarations and rehabilitation reports. We implement data minimisation (processing only what is strictly necessary), pseudonymisation, and processing on European servers under a processor agreement. All processing is documented in a register that aligns with your existing GDPR policy and is available for inspection by the Dutch Data Protection Authority.
Solvency II and model risk management
Solvency II sets requirements for insurers' risk management, including the model risk that AI systems introduce. We document model validation, backtesting results and monitoring procedures in line with EIOPA guidelines. Drift detection, which signals when a model deviates from its original performance level, is built in as standard, so that your actuarial and risk management functions are alerted in good time when a model needs recalibrating.
Bias prevention and fair treatment
AI models that assess risks or calculate premiums must not discriminate on protected characteristics. We test models structurally for bias: are certain postcode areas, age groups or occupational groups disproportionately disadvantaged? The results are reported and, where necessary, corrective measures are taken — fairness constraints within the model, rebalancing of training data or adjustment of features. This is not only ethical but also a requirement under the Wft regarding duty of care and fair customer treatment.
Technology stack for AI in insurance
We choose technology based on the specific challenge, not on platform preference. Typical components in our insurance projects:
Want to learn more about our technical approach? Read about our AI development, our enterprise AI implementation, or explore our custom software services. For insurers also considering a customer app: get an insurance app built.
Why insurers work with Appfront
We are not a consultancy that delivers a report. We are a development team that builds working software and takes responsibility for the outcome.
Sector knowledge as a foundation
We speak the language of the insurance industry: claims handling, underwriting, policy administration, intermediary channels, mandate structures. A solution that does not align with existing workflows and terminology will not be adopted by your staff — and we know that.
Compliance-aware development
The Wft, GDPR, Solvency II and AFM supervision are not appendices to our proposal — they are preconditions of our architecture. We build in explainability, audit trails and bias monitoring from day one. Your compliance department is not the final checkpoint but a stakeholder from the outset.
Open architecture, no vendor lock-in
We build on open standards and give you full access to source code, models and data. If you later wish to change supplier, manage the system yourself or develop it further in-house, nothing stands in the way. Your data and models always remain yours.
Scalable to your organisation
An intermediary firm with fifty employees has different needs than a nationwide general insurer. We tailor scope, architecture and hosting to your scale — from a focused chatbot to an organisation-wide AI layer across multiple core systems. No unnecessary complexity, but also no undersized solutions.
Clear agreements on scope and budget
We work project-based with fixed milestones and a clear scope description. You know in advance what will be built, when it will be ready and what it will cost. No hidden costs, no scope creep without your explicit approval — a way of working that suits the financial sector.
Support after handover
AI models become outdated when data distributions shift, regulations change or core systems receive updates. We offer maintenance agreements that keep your system operational, compliant and up to date — including periodic model validation and performance reporting to your risk management function.
See also our page on custom insurance software development for broader software development in the insurance sector, or read about our enterprise AI implementation approach.
Frequently Asked Questions
The questions below come up regularly among insurers, intermediaries and mandate companies considering AI automation.
Processes that are data-intensive, repetitive and rule-based deliver the most value. In practice, these are: claims triage and initial assessment (FNOL intake, coverage checks, routing), document processing (policy documents, valuation reports, medical statements), fraud detection (pattern recognition in claims history) and customer communication (FAQ chatbots, status updates). Underwriting and acceptance are also suitable, but they require more validation given their direct impact on premiums and coverage decisions. We always recommend starting with the process where the most hours are lost and where data quality is highest.
The Wft requires financial institutions to treat their customers fairly and to be able to explain their decisions. This also applies when AI systems play a role in underwriting, premium calculation or claims handling. We build models with explainability built in: the system not only produces a score or recommendation, but also shows which factors weighed most heavily. This allows an underwriter or claims handler to review and justify the AI suggestion to the customer or regulator. In recent guidance, the AFM has indicated that it expects insurers to be able to demonstrate that AI systems do not undermine the duty of care, and our architecture is designed with that in mind.
In most cases, yes, but the approach differs by system. ANVA offers integrations via their API and file exchanges; CCS has a modular architecture with integration points; Guidewire has an extensive integration framework (Guidewire Cloud Integration). For older or heavily customised systems, we look at structured file exchange (CSV, XML via SIVI standards) or an intermediate layer acting as an API gateway. We always start with a technical exploration: what can the existing software do, what requires custom work, and what is the most pragmatic route? Not every integration needs a real-time API; sometimes overnight batch processing is more effective and more stable.
Bias in AI models is a real risk, particularly in underwriting and premium calculation. If historical data contains systematic prejudices, such as stricter risk assessment for certain postcode areas or occupational groups, the model learns those patterns. We routinely test models for bias using fairness metrics: are protected characteristics, directly or indirectly, weighted disproportionately? The results are reported and, where necessary, we take corrective action: fairness constraints in model training, rebalancing of training data, or removing features that act as proxies for protected characteristics. This is not only ethically sound but also a requirement under the Wft duty of care and the European AI Act, which came into force in 2025.
Computer vision analyses photos and videos of damage (dents in cars, water damage to buildings, storm damage to roofs) and compares them with a database of similar cases. The model can highlight the location and extent of the damage in the image, provide an initial estimate of repair costs and flag inconsistencies (for example, damage that does not match the reported incident). This considerably speeds up the assessment process: the loss adjuster receives a preliminary analysis instead of a blank file. The final assessment and report remain with the adjuster; computer vision provides the groundwork and the supporting evidence.
A focused pilot, covering one process with limited integration, is typically operational within eight to twelve weeks, including the exploration phase, compliance check and user testing. A more extensive implementation with multiple integration points, model validation by the actuarial department and a wider rollout takes longer, depending on the complexity of your IT landscape and the availability of data in usable formats. We always work in iterations, so you see working software early in the project and can steer accordingly. The timeline is often determined not by the technical build but by internal decision-making and compliance approval, and we factor that into the planning from the outset.
Health data falls under the special categories of the GDPR and may only be processed on the basis of an explicit legal ground, which in insurance is usually the performance of the insurance contract or the explicit consent of the data subject. We implement strict data minimisation: the AI model only receives the data strictly necessary for the function. Medical statements are pseudonymised before they enter a language model. Processing takes place on European servers under a data processing agreement, and access to health data is restricted by role. All processing is logged in a record of processing activities. When using external AI services (GPT-4o, Claude), we conclude enterprise data processing agreements and ensure that personal data, and health data in particular, is anonymised before it enters the processing chain.
Considering AI for your insurance processes?
Whether you are a non-life insurer looking to speed up claims handling, a underwriting agency wanting to automate underwriting, or an intermediary seeking to streamline client communication, we are happy to help. It starts with a conversation about your specific processes, systems and ambitions. No sales pitch, just a concrete exploration of what is feasible and worthwhile.
An initial exploration is always no-obligation. We will give you an honest assessment of what AI can deliver in your specific situation.
Get in touch