Companies Race into AI, Yet Adoption Lags, Says KPMG Executive

Companies are rapidly embracing artificial intelligence (AI), yet many still face challenges in demonstrating its tangible value. Mathieu Wallich-Petit, Head of Clients & Markets at KPMG France, shared insights with Euronews Next during VivaTech in Paris.

According to a report released by KPMG in March, 95% of its clients have a well-defined AI strategy, and 64% have already observed measurable outcomes from the technology. However, only 8% can clearly quantify a return on investment. “Our clients do implement a real AI strategy, but in practice, there is still a significant gap,” Wallich-Petit noted.

He emphasized that KPMG’s role is to assist companies in bridging this gap, as AI advancements are outpacing organizational adaptability. “What is remarkable is the exponential pace of technological acceleration,” he said. “However, the adoption within each company tends to be linear.”

The survey revealed that approximately 10% of KPMG clients are currently integrating AI at scale. In the insurance sector, companies are beginning to use AI beyond just claims handling. “Previously, it was mainly about automating claims, but now it spans end-to-end processes, including client scoring, pricing, and customer service,” he explained.

KPMG reports that companies are increasing their AI budgets because boards view the technology as a competitive advantage and a means to attract talent. At the same time, businesses are paying closer attention to whether these investments yield clear and rapid returns, according to Wallich-Petit.

Tips for Companies

The French executive highlighted that the primary focus for leaders transitioning into AI should be on their workforce. “My perspective is that it's fundamentally about people, not just technology,” he said. “Upskilling and training employees is likely the most crucial strategic element in making an AI strategy successful.”

For companies caught between pilots and full-scale deployment, Wallich-Petit advised focusing on embedding AI into daily business processes. “The key is to move from proof of concept and piloting to truly integrating it into operations,” he said.

This also involves stronger governance, improved data management, and more training for employees, according to Wallich-Petit. “We always say it’s about having people in the loop. I think it goes further than that. We need to have people driving with AI,” he added.

AI Sovereignty

Wallich-Petit also pointed out that AI sovereignty is becoming a more pressing issue for companies, especially as businesses rely heavily on a limited number of powerful model providers. “The main theme is not to depend solely on one model, but to have a diversity of models,” he stated.

This concern has become more concrete as access to certain advanced AI models becomes entangled in geopolitical tensions. In May, KPMG and US AI company Anthropic announced a global alliance to integrate Claude into KPMG’s client delivery platform and provide its global workforce with access to the AI assistant.

Just weeks later, Anthropic was ordered by the US government to suspend access to its Fable 5 and Mythos 5 models for any foreign national.

For more details on this story, watch the video in the media player above.