AXA Secure GPT: Generative AI arrives at Axa

Axa is moving to the production phase in the field of generative artificial intelligence. The insurer has indeed announced the deployment of its own service, Axa Secure GPT.

Developed on Azure and incorporating OpenAI’s AI technology, it is at this stage proposed to the 1,000 employees of the AXA Group Operations division (AXA GO). The entity founded in 2019 includes support functions, including IT.

An LLM in the technological “heart” of Axa

Axa Go is thus presented as the technological “heart” of the company and the engine of its digital transformation. Its staff is therefore the first to access the generative AI tools delivered by Axa Secure GPT.

“AXA employees will be able to use AXA Secure GPT to generate, summarize, translate and correct texts, images and codes,” details the insurer in a statement. Axa, a regulated player, claims a secure and compliant cloud operation in terms of privacy.

Axa Secure GPT is currently in phase one of its deployment. However, the company is already planning to extend its use to the group’s 140,000 employees over the coming months.

Secure GPT will soon be an everyday tool

“AXA Secure GPT will soon be a tool in our daily work,” confirms Alexander Vollert, Axa’s chief operating officer and managing director of AXA Group Operations. The use of the solution must in particular prevent the use of public applications – including the OpenAI API.

“The use of open tools can lead to serious problems, including data leaks, security breaches and the loss of intellectual property,” says the Axa executive. The launch of Axa Secure GPT is therefore part of a strategy that is both offensive and defensive.

“AXA has once again demonstrated its ability to innovate quickly, taking advantage of its cloud-based infrastructure. By doing so, AXA becomes one of the first global insurers to develop such a platform on a large scale while managing potential risks,” welcomes Alexander Vollert.