How to ensure AI helps business

With its capability to analyse vast amounts of data in a short time for the insurance industry, Artificial Intelligence (AI) is an increasingly important backbone for insights. In addition, it's decision-making capabilities can be a giant step for business efficiency and new solutions. However, the decisions AI takes must be compliant with applicable laws and regulations, and must also be ethical, unbiased and sustainable. And here lies the crux: to be beneficial to the insurance industry and society at large there are many governance, organisational and cultural conditions that applications of AI need to fulfil. Antonio Grasso and I have put together ten recommended guiding principles we consider to be important for a responsible AI fit for business success.

1. AI applications should protect human dignity, rights and fundamental freedoms. Create an AI policy that ensures compliance with requirements including driving fairness and transparency.
2. Ensure AI governance that is consistent with laws and regulations. Including responsibilities and frameworks to assess and review them throughout the entire life cycle of AI, e.g. with model monitoring.
3. Mitigate the risks of AI applications – by setting up an adequate analytics and an AI risk management framework and related processes.
4. Internal and external transparency to the extent permissible under applicable laws and regulations. Foster explainability, where applicable to help stakeholders make informed decisions while protecting privacy, confidentiality and security. Provide options for recourse.
5. Solid data protection standards, cyber security, data foundations and standardised systems are paramount. Consent to use the data, data quality and quantity are among the key factors to succeed with AI.
6. Clarity on where and how AI in combination with human processes has a positive impact on the value chain – be it to increase efficiency or enable new solutions and on its costs.
7. Understanding the limitations of AI in the customer journey. Human input might remain invaluable in some critical decisions to ensure digital trust.
8. Foster and commit to upskilling the existing workforce on the use of new analytics and AI technologies.
9. To ensure that the ongoing validation of the algorithms and adjustments is performed independently, an additional independent internal or external control function is recommended.
10. The use of AI will likely have profound effects on the insurance industry and society. Therefore, it is necessary to ensure a continuous dialogue with all stakeholders to be able to respond to changing needs and views.