Artificial intelligence may be advancing at an unprecedented pace, yet behind the excitement surrounding new models and capabilities, the conversation at NEXUS Luxembourg 2026 centred on a far more fundamental issue: trust.

During a lively discussion entitled “AI is Popping Everywhere: AI on Demand and Europe’s AI Ambition,” three key voices from the European AI ecosystem explored how the continent can transform its scientific excellence into large-scale adoption. The panel brought together Heli Harrikari, Head of EU Strategy at DIMECC, Cécile Huet, Head of Unit for Robotics and Artificial Intelligence Innovation and Excellence at the European Commission, and Laurent Pulinckx, CIO of Creos Luxembourg. Their shared conclusion was clear: Europe already possesses many of the ingredients needed for AI leadership. The challenge now is connecting them.

Europe's Strengths in the AI Race

Opening the discussion, Heli Harrikari framed the debate around a question that is increasingly shaping technology policy across the continent: how can Europe remain competitive in a rapidly accelerating global AI race while preserving its values and technological sovereignty?

For Cécile Huet, the answer begins with recognising Europe’s existing strengths. “Europe has several strengths and assets on which we must capitalize,” she said, pointing to world-class research, engineering expertise, a vibrant innovation community and a commitment to values that Europeans are unwilling to compromise.

But talent alone is not enough. Cécile Huet highlighted the European Commission’s AI Continent Action Plan, which seeks to bring together the infrastructure, skills, funding and regulatory frameworks needed to create a truly European AI ecosystem.

From EuroHPC supercomputing resources and AI factories to cloud infrastructure, industrial data access programmes and AI skills initiatives, Europe is steadily assembling the building blocks of a comprehensive AI strategy. The goal is not merely to develop AI technologies, but to ensure that Europe remains sovereign in critical digital capabilities while reducing strategic dependencies.

The ambition is particularly focused on sectors where Europe already holds strong industrial positions, including aerospace, automotive, robotics and pharmaceuticals. “We want to keep the leading edge in these sectors in bringing AI,” she explained.

The Conversation Has Changed

While Europe’s policymakers focus on building capabilities, enterprises are undergoing a profound shift in how they evaluate AI. According to Laurent Pulinckx, the questions companies ask today are radically different from those being asked just a few years ago. “Two or three years ago, the discussion centred on comparing AI models and platforms. What is the difference between Copilot and ChatGPT? Which model performs best?” he recalled. Those debates, he argued, have become secondary: “The subject today is where my data are located, who can access it, can I use the result of the AI algorithm, is it trustworthy, is it explainable?”

For organisations operating critical infrastructure, these concerns are far from theoretical.

As the operator responsible for electricity and gas distribution across Luxembourg, Creos cannot afford uncertainty. Innovation is important, Laurent Pulinckx acknowledged, but reliability and resilience are non-negotiable. This reality is pushing enterprises beyond experimentation and towards a more rigorous evaluation of AI technologies. The shift is no longer about identifying the most powerful model; it is about identifying systems that can be trusted in production environments. “The shift has really been moving from a technology point of view to a question of trust,” he said.

Connecting Europe’s Fragmented AI Landscape

One of the recurring themes throughout the discussion was fragmentation. Europe boasts exceptional research institutions, innovative startups and strong industrial players. Yet these communities often operate in parallel rather than together. This is precisely the problem that the AI-on-Demand platform aims to address.

Cécile Huet described the initiative as a bridge between research, innovation and deployment. The platform provides access to European research outputs while simultaneously showcasing commercial AI solutions developed across the continent. Importantly, the platform seeks to do more than catalogue technologies. It aims to help users discover trustworthy solutions and accelerate their deployment.

“The challenge is no longer whether Europe can innovate in AI,” Heli Harrikari observed. “The question is whether Europe can connect innovation with deployment at scale.” The AI-on-Demand platform includes several mechanisms designed to strengthen those connections. A Business Navigator helps startups, SMEs and AI developers gain visibility at a European level, while a Marketplace allows companies to promote AI products and services to potential users.

The platform is also designed to work closely with Europe’s network of Digital Innovation Hubs, creating channels through which AI solutions can reach regional ecosystems across the continent. For Cécile Huet, ecosystem-building is ultimately Europe’s greatest strength. “Europe’s best asset is when we work together, when we join forces,” she said.

Stability Matters More Than Speed

If trust is the foundation of adoption, stability may be its most practical requirement. Laurent Pulinckx stressed that enterprises deploying AI into operational environments need confidence that technologies will remain available, maintainable and certifiable over the long term. For a utility operator, AI is not simply another productivity tool. It may eventually influence systems connected to physical infrastructure such as electricity transformers and network operations.

That creates a very different set of expectations for AI deployment. Companies need solutions that are explainable, manageable and accessible. More importantly, they need confidence that those solutions will still be viable years after deployment. “Structuring the ecosystem is key to accelerating adoption,” he argued. Without that stability, organisations may hesitate to move beyond pilot projects regardless of the potential benefits.

The Role of Regulation in Building Trust

Perhaps the most revealing moment of the discussion came during the closing question: What could become Europe’s true competitive advantage in AI over the next year?

While many might have expected references to supercomputers, research excellence or industrial expertise, Laurent Pulinckx chose a different answer: “the right level of regulation.” His answer touched on one of the most debated issues in global AI development. Europe is often criticised for regulating emerging technologies more aggressively than other regions. Yet Pulinckx argued that carefully designed guardrails could become a differentiating strength rather than a weakness. Trust, he insisted, is what enables innovation to scale. Regulation, when proportionate and practical, helps create that trust. “The right level of regulation is what's going to help to build the trust,” he said. 

Cécile Huet echoed the importance of trust, although from a broader ecosystem perspective. She argued that Europe’s opportunity lies in bringing developers and end users closer together, creating solutions that directly address real-world needs while fostering confidence among all participants.

The Next Phase of Europe’s AI Journey

As the session drew to a close, Heli Harrikari offered a concise summary of the discussion. “Europe already has many of the ingredients needed for AI leadership,” she said. “We have talent, we have research, we have industrial capability and strong values. The real opportunity now is to connect these strengths into trusted ecosystems that make AI easier to discover, adopt and deploy at scale for all of us. And that's the vision behind the AI on Demand platform.”