ERGO AG – How to teach AI read and understand insurance documents

June 13th, 11.15 – 11.45 am

ERGO as many other primary insurances in Germany receives over 100 million pages of insurance related documents from customers every year. Nearly half of it is still coming in via post mail or fax. We are showing how we trained an AI on the previously introduced ERGO AI Factory (Felix Wenzel in 2022) that can classify and extract relevant data from all these documents. The algorithm is supe-rior to the currently used rule based systems and was implemented our highly regulated insurance processes.

Dr. Sebastian Kaiser, Head of Machine Learnig | ERGO AG

Ergo_Dr.-Sebastian-Kaiser_CFP-squareAfter my Diploma and PhD in Statistics, I worked for various companies including consultancy, banking and automotive. In 2017, I joined Munich Re Group driving AI and IoT use cases inside the insurance industry. Since 2021, I am heading the machine learning team of ERGO where we devel-op machine and deep learning algorithms and bring them into production. I have a strong passion for statistics and data analysis including AI.




Dr. Jennifer Betz, Head of AI Advisory | ERGO AG

Ergo_Dr.-Jennifer-Betz_CFP-squareAfter my master degree in economics, I did my PhD at the chair of statistics and risk management in Regensburg. My thesis is related to applied statistics in the area of credit risk modelling. Although I spend a short post-doc period, I decided to leave academia and joined ERGO in 2019 as a data scientist. Actually, I am Head of AI Advisory, leading a team of passionate data scientist. No matter if it is about statistics, machine learning or AI – I am generally enthusiastic about data.