Generative AI for Synthetic Actuarial Data

Introduction & Programme

Access to realistic, publicly available datasets is a significant barrier to advancing the development of assets for insurance analytics and actuarial research. A substantial amount of data in the insurance industry is confidential and proprietary, making it challenging to validate new methodologies. Even when data owners are willing to share anonymised datasets, the effort required to mask the data and navigate bureaucratic processes can often prevent disclosure. This environment limits opportunities to measure progress in the field.

To address the scarcity of real insurance data, there is an opportunity to develop simulated datasets that capture features similar to those observed in practice, thereby reducing friction in data disclosure for development and research. Although synthetic data generation is an active area within the broader machine learning community, its specific application to insurance data remains underexplored.

This web session is designed to demonstrate how to build synthetic insurance datasets using various generative models.

Several models will be employed, starting with Gaussian Mixture Models and extending to the conditional format for the typical non-life insurance dataset used to model frequency and severity. Results will be compared with those of the Conditional Diffusion Model. The seminar will explore the Variational Autoencoder and Generative Adversarial Networks using other insurance datasets. Lastly, we will also see the use of Large Language Models to generate synthetic data by exploiting prompt engineering. 

The data generation process will be conducted by running several trials: reproducing the full number of records in the data fed into the generative models, applying data augmentation, and then omitting sensitive variables to preserve privacy.

Several techniques will be used for the validation of the generated datasets:

• Consistency Tests: to verify that synthetic insurance records strictly adhere to industry business logic.
• Kolmogorov-Smirnov Test: to ensure that the generated data maintains the underlying statistical properties of the original datasets.
• Data Visualization: using univariate analysis, correlation matrices, and other spatial techniques.
• Actuarial Modelling: evaluating predictive utility also through feature importance comparisons.

Preliminary Programme
Tuesday, 10 November 2026
09:00-11:00 Introduction to Gaussian Mixture Models and Diffusion Models. Applied use cases with notebooks.
11:00-11:30 Break
11:30-13:30 Introduction to Large Language Models, Variational Autoencoders and Generative Adversarial Networks. Applied use cases with notebooks.

All the above times are given in CET (Central European Time).

Learning Objectives & Approach

This session is designed for practitioners who seek to understand when generative models add value, how to implement them, and the trade-offs involved. Rather than focusing purely on theory, the session emphasises intuition, hands-on practice, and pragmatic, real-world decision-making.

Participants

This web session is designed for actuaries, actuarial students, data scientists, and product developers working within the risk analytics and insurance sectors. It is particularly relevant for professionals seeking to improve their expertise in applying Generative AI to generate synthetic data.

Attendees should have basic knowledge of Machine Learning, Deep Learning, and Python. Participants will also need active access to Google Colab and a GenAI platform (such as ChatGPT, Gemini, or Claude).

Technical Requirements
Please check with your IT department if your firewall and computer settings support web session participation (the programme Zoom will be used for this online training). Please also make sure to join the web session with a stable internet connection.

Lecturers

Claudio Giorgio Giancaterino
Claudio Giancaterino is a qualified Actuary who works during the day with Intesa Sanpaolo Assicurazioni, an Italian Insurance Company based in Milan. In his free time, he engages in independent AI & Data Science activities. Previously, he served as an assistant professor of Insurance Statistics at the Catholic University of Milan. He collaborated with the IAA and the IFoA on several working parties and participated in data science competitions. He holds workshops and talks at several conferences and meetups. He’s a member of the Astin Actuarial Group and the Italian Actuarial Body Association.

Language & CPD Credits

The language of the web session will be English.

CPD Credits
For this web session, the following CPD credits are available under the CPD scheme of the relevant national actuarial association:

• Austria: 4 points
• Belgium:  4 points
• Bulgaria:  6 points
• Croatia: individual accreditation
• Czechia:  4 hours
• Denmark 4 credits
• Estonia:  4 hours
• Finland: 4 points
• France:  24 points
• Germany:  4 hours
• Greece: 5 points
• Hungary:  4 hours
• Iceland: 4 credits
• Ireland: 4 hours
• Italy: individual accreditation
• Latvia: 4 hours
• Lithuania: 4 hours
• Netherlands:  approx. 4 points (individual accreditation)
• Norway: 4 points
• Poland: 4 hours
• Portugal: 4 hours
• Serbia: 4 hours
• Slovakia:  individual accreditation
• Slovenia:  individual accreditation
• Spain: CAC: 4 hours, IAE: 4 hours
• Switzerland:  individual accreditation
• USA: SOA (Section B): up to 4.8 hours

No responsibility is taken for the accuracy of this information.

Fees & Registration Details

Early Bird Registration Fee (until 29 September 2026):

• For private customers in the EU: €320.00 + VAT of the billing country (example Germany: €380.80 incl. 19% VAT)
• For private customers outside the EU: €380.80 (incl. 19% VAT)
• For businesses within the EU (excl. Germany, with valid VAT ID): €320.00 (net, reverse charge applies)
• For businesses in Germany: €380.80 (incl. 19% VAT)

Regular Registration Fee (from 30 September 2026):

• For private customers in the EU: €420.00 + VAT of the billing country (example Germany: €499.80 incl. 19% VAT)
• For private customers outside the EU: €499.80 (incl. 19% VAT)
• For businesses within the EU (excl. Germany, with valid VAT ID): €420.00 (net, reverse charge applies)
• For businesses in Germany: €499.80 (incl. 19% VAT)

Important VAT Information:

• For private customers with a billing address in an EU country: VAT will be charged at the applicable rate in the country of the billing address. The final amount, including VAT, will be calculated upon invoicing.
• For customers with a non-EU (third country) billing address: Only a non-company billing address is accepted for VAT compliance reasons. 19% VAT applies to all non-EU private customers.
• For businesses within the EU (excluding Germany), Iceland, Liechtenstein, Norway, Switzerland, and the UK with a valid VAT ID: The reverse charge mechanism applies (net price; VAT will not be charged). Please ensure your valid VAT ID is entered correctly during registration.
• For all customers with a billing address in Germany: 19% VAT applies.

Please submit your registration using our online form below. Closer to the event, you will receive further login details to join the web session.

Your registration is binding. Cancellation is only possible up to 2 weeks before the first day of the event. If you cancel later, the full participation fee is due. You may appoint someone to take your place but must notify us in advance. EAA has the right to cancel the event if the minimum number of participants is not reached.

We will send you an invoice via email. Please allow a few days for handling. Please always give your invoice number when you effect payment. All bank charges are to be borne by the participant.

Registration is open until two working days before the web session. If registration has already been closed for this web session, please call us or send an email to contact@actuarial-academy.com in order to find out whether a late registration is still possible.