In a recent partner webinar with Roland Berger’s Dr. Ulrich Kleipaß and Dr. Clemens Frey, the critical role of AI governance tailored for insurers was explored. Here, we recap some of the important learnings from our partners and invite you to watch the on-demand session to learn more.
Why is AI governance necessary? AI governance ensures quality, mitigates risks, and facilitates scalable and compliant AI adoption in insurance, addressing challenges like bias and regulatory demands.
What makes AI governance successful? Key elements include principles, structures, and processes, with success relying on strategic alignment, robust risk management, and integration with existing governance frameworks.
How can AI governance be implemented? AI governance can be implemented through baseline assessments, framework design, stakeholder alignment, and iterative rollouts, supported by tools and external expertise.
Dr. Frey: We are seeing an increasing spend on AI in the financial services industry, both insurance and banking. For insurance, AI is important because it creates more quality, it creates more individualization and, most importantly, more process efficiencies. There are a lot of applications of AI that make it necessary to manage quality and risk across the value chain.
Dr. Frey: First of all, a systematic risk and quality management for all AI applications is needed and for that you need a framework of responsibilities throughout the project. Secondly, this should be based not only on organizational structures, but also on solid statistics, to assess the quality of an enormous number of predictions. Thirdly, you need a common scheme, a common definition of quality across all the applications.
Dr. Kleipaß: There are some clear benefits associated with regulating AI governance, and we see that in four dimensions. One is that you really have this strategic alignment of the different initiatives. It should also give you efficiency gains because there are certain tools and processes that can be used more centrally and really help you scale. Moreover, it has to do with risk mitigation and compliance - if you have it too decentralized and no proper AI governance on it, this might go out of hand. Last but not least AI governance can really help you to also drive cultural change and identify where you need to work on.
QuantPi and Roland Berger offer a comprehensive solution for responsible AI adoption. QuantPi provides technical expertise in scalable quality assessments, ensuring robust, unbiased, and safe AI systems. Roland Berger delivers strategic insights and industry-specific knowledge, guiding the integration of AI governance frameworks within organizational structures. Together, a holistic solution combines technical rigor with strategic alignment, enabling organizations to harness the power of AI responsibly and effectively.
Dr. Ulrich Kleipaß is a Senior Partner at Roland Berger and has been with the firm since 2008. He is responsible for digital projects in the insurance sector. Dr. Kleipaß has previously worked for Allianz, Ergo, and Munich Re and has been involved in projects on AI and AI governance for two years.
.png)
Dr. Clemens Frey is a mathematician with 15 years of experience in consulting, he has a background in risk management, having worked at Munich Re Risk Management. Dr. Frey has been a Partner at Roland Berger for two years focusing on the the transformation triggered by AI and Cloud.