Generali France rides the wave of Microsoft 365 Copilot and agentic AI with Microsoft

Savings, health and pension plans, personal and property insurance … Generali France has established itself as a key player in the Insurance sector in France. The organization’s vision is to extend its value proposition beyond delivering products and services, emphasizing a people-centric approach and the implementation of innovative solutions to support its customers at every stage of their life journey. 

This philosophy is fully embedded in the company’s initiatives, as demonstrated by “Boost 2027,” the three-year strategic plan centered on people, innovation, technology, and data—with AI and generative AI playing a central role. Emmanuel Néré, Director of Innovation and Transformation, and Zineb Taleb, Head of the GenAI Strategic Program, share their insights.

Becoming the tech benchmark in insurance 

While Generali France has leveraged AI for many years—particularly for fraud detection and risk scoring—the advent of generative AI has shifted the paradigm, prompting the insurer to scale up and go one step further. “This technology brings us into a new dimension,” observes Emmanuel Néré, “one that allows us to enhance the client and distributor experience and improve the quality and personalization of our offerings.” 

The AI component of “Boost 2027” was designed with this in mind, through the democratization of tools for all, the implementation of enterprise-focused use cases built on the group’s expertise, and the experimentation with agentic AI via the development of several AI companions. “We aim to become the Tech benchmark in insurance by leveraging what technology can offer our clients and distribution networks, creating seamless and efficient solutions,” he adds. 

From Microsoft 365 Copilot to Copilot Studio: leveraging powerful tools to achieve their ambitions

The group brings these ambitions to life by drawing on the expertise of Microsoft, a long-standing partner. “We have been using Microsoft products for a long time; our decision is part of a desire to maintain continuity in the user journey,” explains Emmanuel Néré, adding: “Microsoft also provides in-depth listening and expertise to help us advance.” 

Generali has given every employee the opportunity to access generative AI. 100% of employees now have access to Copilot Chat, and 3,700 of them benefit from Microsoft 365 Copilot to enhance their daily office tasks. “70% of our workforce have adopted the application in their roles, generating an average of fifteen prompts per user per week,” notes Emmanuel Néré. Leading use cases include summarizing email threads or information, structuring or producing documentation or communications, and retrieving precise information from particularly rich bases and document repositories. The group’s Cognitive Factory, the group’ operational unit of excellence dedicated to automation and AI, is accelerating with 17 business use cases in production across the company and nearly thirty more planned by the end of 2025. 

Meanwhile, Microsoft Copilot Studio and Azure OpenAI are among the latest solutions added to this arsenal. The goal? Empower teams to become “augmented employees” through the development of agents. “Copilot Studio integrates easily into our ecosystems and information sources,” says Emmanuel Néré, “the agents we create with this tool truly help employees manage their operations.” About fifty agents have been deployed to address more specialized cases. 

Collaboration with Avanade has been instrumental in integrating AI solutions, fostering adoption, scoping, and efficiently deploying the first use cases.

AI Agents: dedicated support for experts and clients 

From extracting information from unstructured data and running hyper-personalised marketing campaigns, to assisting with content creation and standardising responses to requests for proposals (RFPs), at Generali France, everyone is embracing generative AI. "Our employees no longer need to perform repetitive, low-value tasks; they can refocus on their expertise, their specific skills, and human interaction," notes Emmanuel Néré. 

Take, for example, helpdesk operations, where employees are now supported by a constellation of agents working collaboratively, in solving cases. Some agents assist with drafting intervention requests, others track case progress, while others manage case escalation to identify the need for specialist expertise. Microsoft solutions have also powered the development of a 24/7 voice assistant, capable of reassuring claimants before a human steps in. The result? Nearly 1.3 million calls—representing 30% of requests—are resolved directly by clients, with no human intervention needed. This is a real boon for teams whose quality of work life conditions improve, especially during peaks of climate-related claims, and for clients who benefit from faster compensation. At the same time, the home insurance subscription process has been simplified: instead of answering around thirty questions, clients can now hand over to an AI agent who manages the process on their behalf. 

Compliance, a key issue in the insurance sector, is not left behind. Here, AI enables a 360° visualisation of clients and their networks, to meet legal obligations regarding beneficiary searches. The results speak for themselves: in 2016, 15,705 contracts were classified as unclaimed, compared to just 802 in 2023. In 2017, 33% of new deaths were identified and processed within a year, a figure that rose to 88% in 2023. 

Generative AI is also being harnessed for simple claims management: in 2024, 2.1 million operations were processed by RPA robots. Notable examples include the analysis of cracks on walls during drought episodes, 10% of which now require no human intervention, and the automatic recognition of more than 15 types of documents, enabling the automation of the qualification of supporting documents sent by clients and ensuring their completeness. "On a daily basis, we work with nearly 300 robots responsible for automating 11% of our processes," explains Zineb Taleb, "and while they are efficient, their capabilities can be limited, which is why we have added a layer of generative AI." The result is a shift from automation to the empowerment of teams. 

Generali now tops the Net Promoter Score rankings, which recognise companies excelling in customer satisfaction. "The generative AI programme has helped support this quality of service and its positive perception among clients," says Emmanuel Néré. 

Simplicity, immediacy, personalisation: the pillars of tomorrow’s insurance 

"While clients' insurance needs remain unchanged, agentic AI is interesting in that it brings simplicity, immediacy, the ability to synthesise data, and a personalised relationship," says Emmanuel Néré.

On the insurer side, technology frees up experts’ time while giving them rapid access to the right information: 4.5 million emails have been read and indexed by agents to provide greater insight into clients’ intentions. 

Experts themselves are fully convinced by this technological revolution. "70% of our employees believe that AI will transform their profession in the coming years," warns Emmanuel Néré, "change management is being carried out with them, and through sustained social dialogue within Generali France." The impacts are manifold: enhancing expertise, boosting efficiency and collective performance, enriching the user experience to retain and develop talent. 

Building a federative, controlled and ethical technological approach 

In this context, support programmes targeting all business lines have been launched, from "prompt bars" and "office coffees" to training led by the Generali Academy. Not to mention the deployment of technical teams by Microsoft, and the awareness and adoption efforts from 300 company ambassadors. "We're not here just to do business. Our mission is to invent the insurance of tomorrow," insists Emmanuel Néré, "the group is moving in this direction and we are very proud of it." 

Generali France also places a premium on practising ethical and responsible AI, fully aligned with the AI Act. "This is inseparable from our innovation approach. We carefully analyse our various use cases, ensure the sustainability and carbon footprint of our models, just as we pursue a strict procurement strategy with our suppliers," explains Emmanuel Néré. 

Algorithm training is also strictly governed to prevent any potential drift, and infrastructures are duly protected. "Insurance is based on trust, so securing data and applications such as those from Microsoft is fundamental," he concludes.