The European Forum for Risk and Insurance professionnals
Okay Gunes, Head of AI and Data Sciences, Arondor
Why is participating in an European risk management important for you?
identifying the common and uncommon factors between risk and uncertainty in decision making represents 10 years of my academic research. I think risk management community, and more precisely European risk management, allows me to share and deepen my knowledge in risk management.
What does “Aim for the future” mean for you?
Aim for future is to increase the usage of AI solutions in Risk Management.
I want to develop my answer a little bit within the context of AI applications and Risk Management.
Today, even if AI is already used by certain companies for detection-prediction or identification, it is not quite convincing that AI is stands for making a better decision than those done by human being. The main reason is that the costs of making an erroneous decision are still very high in certain sectors.
However, for the risk management this is not the case. AI can mainly be used for understanding the patterns hiding behind the data or further to organize, capture and identify the valuable information in order to better optimize any processes of Risk. Here,we can’t really talk directly about costs of decision! Mostly AI provides new or ignored information in more effective way by increasing the quality of Risk Managers decision.
It is inevitable that one day AI applications enters in Risk management processes. Thus, my answer totally depends on how much Risk managers are ready for changing and using data driven AI solutions. In my opinion, first step is to increase vision and to eliminate attitude of Risk Managers against usage of AI in their tasks. Hence, in the white paper, we tried to clarify as much as possible the best attitude for integrating AI in Risk management.
You will be presenting in the “From data to AI: the new scope of the risk manager in 2020”, what would you like the audience to remember after your session?
The importance of data collection, the limits of AI solution and the role of collaboration.
In fact, basic understanding of the data prevents from making elementary mistakes. Data collecting requires involving both data scientists, data engineers and domain experts together in collection plan in order to ensure that gathered data is suitable for the model to be built. What mostly happens in real life is that the data has already been collected before the data scientist arrives. Remember that data scientist tries to make the best of what is available.