AI in France: 50% of Projects Fail Due to Lack of Governance

⚡

Key Takeaways

1A study reveals that 94% of French companies have launched AI projects, but 50% fail to scale them.

2The lack of data governance is identified as the main cause of failures in AI integration.

3Companies need to recruit data governance experts to successfully transform with AI.

💡Why it matters — Without strong data governance, companies risk costly failures in their AI projects, jeopardizing their competitiveness.

A Massive Yet Incomplete Adoption of AI in France

According to a study conducted by SalesForce and OpinionWay, 94% of French companies have initiated artificial intelligence projects. However, an alarming observation emerges: 50% of these initiatives fail to move beyond the testing phase. This paradox highlights the challenges faced in scaling these projects, despite a clear desire for digital transformation.

The Underpinnings of AI Adoption

Behind these impressive figures lies a more nuanced reality. Jonathan Bodin, strategy director at Seenovate, points out that most companies claiming to use AI are actually practicing "Shadow AI." This means they are using AI in an isolated and uncoordinated manner, often without the oversight of the IT department. In reality, only 20% of companies truly integrate AI into their processes.

AI as a Revealer of Organizational Weaknesses

The introduction of AI in companies often merely highlights existing dysfunctions. These issues include unclear responsibilities regarding data management, siloed teams, and inefficient decision-making processes. Élise Tissier, director of Bpifrance Le Lab, asserts that a well-defined data strategy is essential to fully leverage AI.

The Risks of AI Without Governance

Olivier Marcheteau, CEO of Freelance.com, emphasizes that leveraging AI relies on a solid data foundation, specific technological skills, and a revision of business processes. Without this, companies expose themselves to tangible operational risks. He notes that while AI can achieve 97% accuracy, the 3% inaccuracy can lead to costly consequences.

Emerging Roles for Successful Transformation

To overcome these challenges, companies must look towards new professional profiles, such as data governance experts, Data Stewards, AI Ethics Officers, and project managers trained in AI. These roles, at the intersection of technology, business, and compliance, are crucial for sustainable AI deployment. They are not a luxury reserved for large corporations but a necessity in an increasingly demanding regulatory environment, particularly with the AI Act.

An HR Challenge and a Clear Message for Leaders

The challenge regarding human resources is real. According to the barometer, 62% of companies believe they lack the internal skills necessary for their AI projects. Although the market is responding with a 273% increase in AI-related job offers between 2019 and 2024, the alignment between needs and available profiles remains insufficient.

For AI to become a true performance opportunity, leaders must be willing to rethink their processes and treat data as a strategic asset. The success of AI projects depends on effective data governance, supported by skilled individuals and committed executive leadership. The algorithm is ready, but the question remains: is your organization?