Introduction
Artificial Intelligence(AI) is no longer a promise for the future of banking, it is already a concrete driver of transformation. After the technology sector, the banking industry is expected to experience the largest productivity gains from AI, particularly through generative and agentic AI.
Client
BNP Paribas Fortis
Client since
2018
Services
Solutions
Technologies
The problem
At BNP Paribas Fortis, the AI Tribe functions as an internal scale-up, delivering three key services to the bank, each supported by a flagship use case:
- Client Experience - Powered by "Samy", the virtual assistant enhancing customer interactions
- Augmented Employee - Supported by "Yara", the internal generative AI assistant for employees
- Automation - Driven by strategic partnerships, such as with Mistral
The ambition is clear: minimize administrative workload to focus on human interaction with customers. However, unlocking the value of AI requires more than technology alone. As the impact of AI in banking grows, trust becomes essential. For BNP Paribas Fortis, one principle guides this transformation:
“There is no AI without responsible AI.” Anthony Belpaire, Tribe Lead AI at BNPP Fortis
The emergence of the European AI Act made this challenge even more pressing. Financial institutions must ensure that AI systems are transparent, explainable, auditable, and subject to human oversight. At the same time, innovation cycles in AI move faster than ever, and the bank needed a way to balance speed with governance.
Without a strong AI Governance framework, AI initiatives across the bank may become fragmented, leading to inefficiencies and compliance challenges. BNP Paribas Fortis needed a single governance process that would allow the organization to innovate quickly while integrating responsible AI principles from the start.
How we solved it
.png)
A collaborative approach with stakeholders
To address the challenge of responsible AI, BNP Paribas Fortis brought together stakeholders from across the organization to design a governance approach that would set the tone and enable AI to scale safely and responsibly, involving people from Data & AI, Compliance, Risk & the business
A comprehensive AI risk management process
With the right people involved, the bank implemented a concrete AI risk management process integrated into the AI development lifecycle. Risks are identified early in the design phase, managed throughout the development process, and subjected to a second review before deployment into production. Once an AI system is in production, it is continuously monitored to ensure it behaves as expected and remains compliant over time.
Dual accountability
An Important aspect of the framework is clarifying ownership of AI systems. Every AI system is now assigned both a business owner and a technical owner. BNP Paribas Fortis’ Responsible AI Officer illustrates this with a simple analogy:
"Just as a car requires both a manufacturer, ensuring a safe design, and a driver, accountable for safe driving, AI systems must have both a business and a technical owner to govern their development and usage." Philippe Damoiseau, Responsible AI Officer at BNP Paribas Fortis
The results
By embedding responsible AI into its strategy, BNP Paribas Fortis has the trust and confidence in their model. The central AI inventory gives an overview of the status of each project and highlights important information such as the AI Act classification.
The AI risk management process ensures that the right people are involved at the right time and that actions and oversight is managed in a risk-based manner. What previously could take months can now be assessed and addressed in a single process in a matter of days or weeks.
100+ AI projects
Today, the impact of AI for BNP Paribas Fortis is no longer a distant prospect, but a concrete reality. The Belgium bank develops 100+ AI projects.
70 million euro
These AI projects are already resulting in a value generation of 70 million euro.
270 full time equivalents
Concretely AI is performing the work of 270 full time equivalents, this allows to pivot people to more valuable tasks like customer facing contacts.
Since 2017, Datashift is supporting us in shifting our data practices from a reactive, regulatory approach to a proactive, business-driven capability. Over the years, Datashift has consistently proven to be a trusted long-term partner, helping us shape a clear vision, design effective solutions, and integrate them safely into our operational processes. Our partnership now extends to safeguarding the value of AI solutions through the development of trustworthy AI.
Since 2017, Datashift is supporting us in shifting our data practices from a reactive, regulatory approach to a proactive, business-driven capability. Over the years, Datashift has consistently proven to be a trusted long-term partner, helping us shape a clear vision, design effective solutions, and integrate them safely into our operational processes. Our partnership now extends to safeguarding the value of AI solutions through the development of trustworthy AI.
Key Learnings
More case studies
We deliver impact where it matters most.
Unlocking Viewer Insights with a Scalable Data Platform
Streamz is the Flemish video-on-demand platform offering local and international series and films through a subscription-based model. Streamz focuses on delivering high-quality content tailored to local audiences, supported by a strategic partnership with Paramount.
Building Practical AI Governance at KBC with Collibra
KBC, one of Belgium’s leading financial institutions, has a long-standing focus on responsible innovation and efficient risk management. With the European AI Act introducing new expectations around transparency, risk management, and accountability, the organization saw an opportunity to strengthen and streamline its approach to AI governance. For KBC, this meant bringing different teams’ perspectives on AI together into one coherent framework—without disrupting existing processes.
