“Now it’s much more about accountability. We need to see metrics moving.” — Jean-Philippe Avelange, Chief Information Officer, Expereo

In this Technology Reseller News podcast, Doug Green speaks with Jean-Philippe Avelange, CIO of Expereo, about a major shift now underway in enterprise AI strategy: the move from experimentation to measurable return on investment.

Avelange explains that 2024 was largely a period of AI discovery, while 2025 brought broader rollouts and aggressive investment. In 2026, however, the tone has changed. Boards and executive teams are no longer satisfied with AI pilots and proofs of concept alone. They want evidence that AI is improving cycle times, raising service quality, and producing tangible business results.

Expereo, a global connectivity provider helping multinational enterprises source and manage internet-centric connectivity worldwide, has a front-row view of this shift. The company works closely with enterprises facing both the opportunity and the infrastructure demands created by AI-driven transformation. According to Avelange, many organizations are learning that AI success depends less on adding tools for employees and more on rethinking business processes, governance, and data structures so AI can be embedded directly into how work gets done.

That distinction is central to Expereo’s view of embedded AI. Rather than simply providing copilots or chat interfaces to employees, embedded AI places intelligence inside the process itself, allowing agents, skills, and automation flows to perform work in a more autonomous but still controlled way. This requires much more than model access. It demands clear process design, reliable data ownership, and a disciplined understanding of what good outcomes look like.

Avelange notes that this is one reason many AI pilots fail to reach production. Companies often move too quickly, expecting AI to replace effort without first resolving issues around fragmented data, unclear workflows, and inconsistent governance. In contrast, the AI initiatives that now survive budget scrutiny are those tied to specific use cases with clear, near-term business impact.

Inside Expereo, that means focusing on practical gains in areas such as service assurance, multilingual ticket handling, supplier coordination, and customer response quality. The company began with straightforward generative AI use cases, but is now moving toward more structured, embedded intelligence that helps teams make better decisions faster and more consistently.

Avelange’s advice to enterprises entering this new phase of AI spending is to start with the “boring” processes first: the repetitive, manual, and measurable workflows where governance can be built, data can be normalized, and success can be clearly demonstrated. From there, organizations can expand AI more confidently into broader and more transformative parts of the business.

To learn more about Expereo, visit https://www.expereo.com/.