AI is a natural fit with insurance, where the goal is to assess and predict risk. AI is accelerating AXA’s capabilities in both these areas, enabling them to better understand the world and serve their customers responsibly. AXA is also working with Microsoft to shed light on the ethical and social issues of AI.

Explore AXA’s story
2/5

We want to make sure that the data is bringing value to what we’re trying to predict. Where you’re trying to go for causality, you need to have a very good rationale for what you do. That's how insurance is built.

Marcin Detyniecki Head of R&D and Chief Data Scientist, AXA
Marcin Detyniecki
Head of R&D and Chief Data Scientist, AXA

AI in action

The business of insurance is fundamentally about managing risk. By using data to estimate the likelihood that something will happen, insurance companies like AXA can determine the cost. “For centuries, that’s the way insurance has been working—using data,” says Jérôme Lafon, Head of connected car business domain, Data and Tech Innovation, AXA. AI and machine learning are improving the way AXA can predict risk and deliver services to their customers.

Marcin Detyniecki, Head of R&D and Chief Data Scientist at AXA, cites an example of preparing a call center for an impending snowstorm.

Marcin Detyniecki Head of R&D and Chief Data Scientist, AXA
Marcin Detyniecki
Head of R&D and Chief Data Scientist, AXA

AI enhances AXA’s capacity to:

Improve customer service

AI helps AXA to respond in the proper manner through more accurate and robust prediction capabilities. By increasing preparedness, AXA can ensure more customer service personnel are ready to process claims.

Forming coalitions to make a greater impact

In 2018, Microsoft founded the Impact AI collective, which gathered over 65 members across French industries spanning startups, big corporate organizations, and public entities. The goal? Engage in future policymaking regarding trustworthy AI, including establishing guiding principles.

Marcin Detyniecki, Head of R&D and Chief Data Scientist at AXA, leads the collective’s Responsible AI work group, which is creating best practices for organizations implementing Responsible AI governance. Based on Impact AI’s principle of transparency, AXA is working with the international Organization for Economic Co-operation and Development (OECD) to explain algorithmic predictions related to the causes of economic crises. 

Impact AI:

Seven principles for trustworthy AI

01
01

Dignity

AI serves the user, who makes their own decisions.

Microsoft is committed to making sure AI systems are developed responsibly. Learn more about our principles and approach to responsible AI.

5/5

Learning from AXA

A leader in the digital transformation of the insurance sector, AXA is also working with Microsoft to lead on responsible implementation of AI. They’re combining their efforts to enhance the services they provide to customers with a thoughtfulness of the technology’s impact.

How is your organization bringing responsibility into your AI strategy?

Discover more about AXA