Akur8: the X-AI Insurance Pricing Solution
Akur8 offers a SaaS solution for risk modeling and price optimization designed for actuaries. Using cutting edge machine learning, they have automated GLM modelling. From variables and interactions selection to geographic smoothing, the client decides the tradeoff between clarity and performance. The client can immediately spot anomalies and discover new patterns. The Parish-based startup leverages the latest machine-learning technologies, fueled by proprietary algorithms developed specifically for insurance pricing.
At Akur8, they aim to disrupt insurance pricing using machine learning techniques. They provide an all-in-one cloud-based pricing software to help actuaries, insurance data scientists and decisions makers understand their business: what price should they propose to increase the margin or acquire new clients? Insurance pricing relies heavily on predicting customers behaviors and future:
Will they undergo claims? Will they buy the product at this price? Will they renew their subscription next year? Insurers need to predict the future and act on these predictions.
Therefore, they created an XAI (Explainable Artificial Intelligence) with proprietary technologies that don’t exist anywhere else, specially designed for actuarial modelling. This technology proved to be much more effective than classic actuarial work and provides substantial gains in pricing quality and delivery speed. Better, faster & stronger.
Akur8 is the only machine-learning solution designed for actuaries. It allows to:
- automatically build clear and auditable risk models in a few days instead of months, dividing modeling time by 10
- improve the model performance (Gini index) by 10%
- give the actuaries a full understanding of the models created, and ability to include its own knowledge,
- offers a clear and integrated process designed for Insurance pricing.
How it works
The actuary enters its data in the tool; models are built automatically thanks to the machine-learning algorithms and proposed to the actuary. If the actuary has additional knowledge, it can review, challenge or improve the models created.
- upload the data-base; review the data to check for errors
- set-up a modeling project, tell the software what information on the clients you want to predict – like the number of claims…
- run the algorithm
- review the results: models are presented by the machine
- pick the solution you prefer and review it in-depth, check which variable plays a role in the model, and how it contributes exactly. Modify things you don’t like
- export your final model, to run it in your environment and also export the model’s documentation.
Why we selected Akur8 for DIA Amsterdam
Akur8 resolves the Automation + Performance + Interpretability Dilemma. The recent machine learning revolution seems to have both automated and increase the performance of risk predictions. However, all the trending techniques such as GBM (Gradient Boosting Machines) are black box models: interpretability is completely lost. Building a performant and understandable GLM is now as easy as choosing the range of number of variables: the Artificial Intelligence of Aktur8 will optimally choose the most important variables and automatically tune the effects of the variables.
At DIA Amsterdam Guillaume Beraud-Sudreau, chief actuary of Aktur8 will present the Akur8 risk modeling solution applied to a typical P&C retail demo.
Who is Akur8?
The team of Akur8 is coming from years of operative experience in the insurance market. They believe strongly that it is possible to have Automation, Performance and Interpretability at the same time.
Akur8 is part of Kamet. Kamet is an insurtech startup studio, a company that creates companies in the insurance, protection and assistance space. Basically, Kamet develops ideas, combines them with talents and capital and turns them into successful companies. To enable rapid product development and launch, Kamet provides their startups and teams with the right environment.
Seed investment by Karmet and SeriesA are under preparation.
Guillaume Beraud-Sudreau, chief actuary of Aktur8