Why we went to Paris
To better understand how rapid advances in AI are transforming organisations, a delegation of leaders and founders from the Netherlands participated in a two-day programme in Paris, immersing top executives in the city’s vibrant AI ecosystem.
The goal was to engage directly with leading AI pioneers, including startups, established companies, and research institutions. Here is an overview of the key ideas and takeaways that emerged from that programme.
Understanding AI’s transformative power
Day 1: Building strategic clarity and leadership foundations
Exploring where AI will lead us, Carlo van de Weijer, General Manager of the Eindhoven AI Systems Institute at Eindhoven University of Technology, outlined how the value of human work is shifting from execution to strategic direction (the prompt) and critical evaluation (the check). At the start of the process, the prompt must provide strategic clarity, define the problem, and ask the right questions, while the check applies human judgment, context, ethical considerations, and accountability to the AI-generated output. He warned that the real danger is not that AI will become like humans, but that humans will start to think like computers. The ultimate goal is to use AI to augment, not diminish, human creativity and judgment.
Focusing on the importance of leadership, Jorg Schalekamp, a partner at Deloitte Netherlands, explained that scaling AI beyond experiments requires a clear, integrated strategy that links use cases to measurable business value. As it touches so many dimensions (strategy, organisation, risk, technology, ecosystems), AI should be owned at the board level. Organisations need an operating model and governance that balance autonomy with central coordination to ensure consistency, risk control and reuse. People, culture and capability building are as important as technology — invest in skills, change management and multidisciplinary teams to capture value.
In an unpredictable AI-driven world, Nathan Furr, Full Professor of Strategy at INSEAD, contended that effective leadership now depends on reframing uncertainty as an opportunity. He called on leaders to combine strategic vision with organisational change and a disciplined experimentation engine, while prioritising the development of cognitive and emotional skills that enable people to act, adapt and find upside in uncertainty.
Finally, Joël Belafa, Co-Founder & CEO of Biolevate, outlined how his health-tech start-up is using AI to accelerate the time-to-market for new medicines. Biolevate is dedicated to solving the "documentation bottleneck" in drug development, where the speed of scientific discovery often outpaces the ability to produce required regulatory paperwork.
"Scaling AI beyond experiments requires a clear, integrated strategy that links use cases to measurable business value. "
Harnessing and scaling AI capabilities
Day 2: Execution, implementation and organisational transformation
Addressing how enterprises can best harness AI, Naser Bakhshi, a partner with Deloitte Netherlands, explained why scaling AI should be treated as an enterprise-wide transformation, not a simple IT project. The goal is to move beyond isolated pilots and create scalable platforms to lead a sustained movement.
Building on that theme, Frederike Rip and Roos Erkelens of Deloitte focused on how to quickly bring about the organisational changes necessary for AI adoption. Leaders must transform tensions around work and talent into “triumphs” by redesigning work, creating new human-centric roles (e.g., AI orchestration lead, prompt systems architect), and building "purple teams" with a mix of complementary skills.
To help people acquire the necessary skills, Paris-based start-up Mendo has developed an upskilling platform. Quentin Amaudry, Co-Founder & CEO of Mendo, explained how the work-integrated platform is designed to democratise access to AI training, moving away from passive learning to a "learn-by-doing" model.
Highlighting the importance of the open-source community as a driver of AI innovation, Jimmy Mianne of Hugging Face, described his organisation as the "GitHub of Machine Learning". Hugging Face is on a mission to democratise state-of-the-art AI by providing tools that allow anyone to build, train, and deploy machine learning models. In Jimmy Mianne’s view, the most important breakthroughs happen when technology is transparent, shared, and accessible to everyone.
Making the case for responsible AI, Alix Rübsaam, Vice President of Research, Expertise and Knowledge at Singularity, stressed the direction a technology takes is the result of human choices. It is crucial to understand the limitations of AI and recognise that no algorithm is truly free from bias. A key question to ask is: "How might the data I’m using influence the outcome?"
Finally, Bart Pustjens of Deloitte explained how employing a sovereign cloud can create business value through greater control, as well as help to reduce risks. However, achieving data, technology, and operational sovereignty comes with increased cost and complexity. Bart Pustjens advised organisations to start pragmatically: run governed proofs of concept, prioritise portability and resilience in procurement, and involve the right stakeholders to scale successes.
Key takeaways and next steps
Leadership and vision
- The speed at which value can be generated with AI is astonishing. There are three potential sources of value: productivity gains, enhanced experiences (employees and customers) and capability building.
- AI systems can synthesise an organisation's collective knowledge, breaking down information silos. This forces teams to operate from a shared vision of the future, rather than negotiating based on fragmented views of the present, and focuses leaders on what truly matters.
- The successful enterprise-wide adoption of AI represents a fundamental strategic transformation, not merely a technological upgrade. It demands decisive, top-down leadership to integrate AI explicitly into the core business strategy.
- Leaders need to frame the AI conversation around opportunities and potential, moving beyond fear or risk-aversion. This inspires the organisation and creates momentum. A leader's ability to set a clear vision, establish guardrails, and inspire action is the single most important factor in the success of AI initiatives.
- Success requires a proactive approach to human capital, cultivating a culture of experimentation and developing new core competencies, such as effective prompting, across the workforce.
Execution and capabilities
- Do not treat AI as a standalone IT project. It must be explicitly woven into the fabric of the overall business strategy to guide investment, resource allocation, and priority setting.
- The challenge lies in creating a repeatable, scalable pathway to move successful experiments from isolated proofs of concept into enterprise-wide solutions. That requires robust governance, with clear data provenance to build trust, and commercial models that reflect the new value AI creates.
- In the fast-moving AI landscape, it is sometimes necessary to develop a solution concurrently with selling or deploying it. This requires agile methodologies and a close feedback loop between development teams, clients, and end-users.
- Effective interaction with generative AI is a skill. Champion the development of prompt engineering skills across the workforce, turning practitioners into expert users who can extract maximum value from AI tools.
- Organisations across all industries are grappling with the same fundamental questions about strategy, talent, and governance. There is value in sharing experiences and learning collectively.
