[VIVATECH] AdvanThink et Quandela démontrent la capacité à intégrer l’Intelligence Artificielle Quantique dans des modèles éprouvés de détection de fraude aux paiements  

Paris-Saclay, le 17 juin 2025 

AdvanThink, leader européen de la détection de fraudes aux paiements en temps réel et Quandela, startup française leader dans le calcul quantique photonique montent un partenariat scientifique visant à créer le futur de la détection des fraudes aux paiements.  

La promesse ? En unissant le potentiel des deux technologies, les deux pépites pourront créer des modèles de détection de fraudes appliqués au monde réel qui seront à la fois plus performants et plus efficients. 

La détection de la fraude aux paiements en temps réel est un défi technologique, opérationnel et stratégique majeur, qui nous concerne tous. Face à l’évolution constante des techniques de fraude, AdvanThink développe depuis 35 ans des outils stratégiques basés sur l’Intelligence Artificielle, pour apprendre en continu et s’adapter. Dans cette course à la performance, l’Intelligence Artificielle s’impose comme un outil stratégique, capable de repérer des signaux faibles à grande échelle, avec une précision et une rapidité inédite. 

Mais les exigences de demain seront encore plus fortes : des modèles toujours plus rapides, précis, économes en énergie et résilients face à des attaques sophistiquées. C’est dans cette perspective qu’AdvanThink et Quandela explorent ensemble les apports de l’informatique quantique pour repousser les limites de l’Etat de l’Art de la détection de fraudes.  

Le partenariat scientifique vise dans un premier temps à la réalisation d’une preuve de concept permettant de démontrer la valeur apportée par l’intégration d’un algorithme de Quantum Machine Learning dans les processus industriels d’AdvanThink. 

« Depuis 35 ans, AdvanThink met l’innovation technologique au cœur de sa stratégie de développement. L’IA quantique représente une promesse importante dans la détection de fraudes et représente un levier réel d’innovation pour construire la sécurité des paiements de demain – et les institutions financières doivent se préoccuper dès maintenant de cette transformation. Nous avons déjà réussi à intégrer la technologie de Quandela au sein d’un pipeline AdvanThink avec tous les pré-requis d’un système industriel prêt à être déployé. Ce premier démonstrateur est plein de promesses pour les experts de la lutte contre la fraude. » déclare Brice Perdrix, CEO d’AdvanThink. 

« Quandela a développé un modèle de quantum machine learning améliorant l’évaluation des risques de crédit. Cet algorithme a également un fort potentiel dans la détection des fraudes aux paiements. Le partenariat avec AdvanThink permet l’intégration de ce modèle dans un workflow industriel et la comparaison aux meilleurs produits disponibles sur le marché. » souligne Niccolo Somaschi, cofondateur et CEO de Quandela.

Launching at GTC Paris, MerLin democratizes quantum machine learning by integrating with classical AI tools—backed by GPU-accelerated performance 

Paris, France – June 11^th – Today, Quandela announces MerLin, a groundbreaking quantum computing framework designed for and by AI data scientists. Set to debut at NVIDIA GTC Paris, MerLin redefines quantum machine learning (QML) with a GPU-first approach, enabling researchers to simulate and benchmark algorithms beyond the limits of today’s quantum hardware. 

Quantum Meets AI: A Collaborative Future 

MerLin positions itself as the “quantum layer for data scientists” – contrasting with other quantum machine learning tools that target quantum scientists. By abstracting quantum complexity into familiar workflows (e.g., PyTorch/scikit-learn integrations), MerLin empowers AI practitioners to prototype hybrid quantum-classical models in hours, not months. Early adopters – including teams from the Perceval Quest, and researchers from Mila, NYUAD’s QML Lab and Scaleway – are collaborating with us to leverage MerLin and bridge classical and quantum workflows. 

“Quantum shouldn’t demand a PhD to use,” said Niccolo Somaschi, co-founder & CEO of Quandela. “MerLin gives data scientists a GPU-accelerated gateway to quantum advantage while ensuring their work remains compatible with real hardware today—and tomorrow. By integrating benchmarks and noise-aware validation, we’re addressing a critical gap: the lack of reproducible metrics in hybrid algorithm research.” 

“Powerful simulation tools are essential to develop better algorithms and accelerate the path to broad quantum advantage”, said Sam Stanwyck, Group Product Manager for quantum computing at NVIDIA. “MerLin solves a critical ecosystem need by opening the door for the broader research community to develop with photonic quantum circuits.” 

Key Innovations 

1. GPU-Optimized Simulators: 

• Leveraging NVIDIA CUDA-Q, MerLin delivers high-performance simulation for photonic quantum circuits, enabling tests for hardware that doesn’t yet exist (e.g., 24+ qubit systems). 

2. Benchmark-Driven Progress: 

• MerLin establishes reproducible metrics for hybrid algorithms, addressing the “benchmarking gap” in QML research—where thousands of papers lack standardized comparisons. 

• Integrated with Quandela Cloud, it enables immediate validation of GPU-optimized algorithms on real photonic hardware, studying noise impact and scalability. 

• Targets pragmatic use cases like quantum-enhanced kNN, GANs, and variational algorithms—backed by hardware-aware compilation. 

3. Photonic-First, Future-Proof: 

• Designed for today’s photonic QPUs (e.g., Perceval-based systems) but architected to adapt to next-gen hardware. 

• Features like dynamic circuit recompilation ensure code scalability across hardware generations. 

Who Uses MerLin? 

• AI/ML Practitioners: Prototype quantum layers without rewriting classical pipelines. 

• Quantum Researchers: Access photonic-specific tools (e.g., boson sampling) with GPU-accelerated simulation. 

• Enterprises: Pilot hybrid quantum-AI workflows with clear ROI benchmarks. 

“MerLin allowed us to adapt existing algorithms to a photonic-native format within a short timeframe. The platform offered useful comparative insights that contributed to our ongoing research and publication efforts”, said Dr. Louis Chen, an early user, Research Associate at the Quantum Centre of Imperial College London (Imperial QuEST) and participant in the most recent Perceval Quest.

Availability & Strategic Vision 

MerLin will be freely accessible to accelerate adoption, with enterprise tiers for advanced features. The roadmap includes: 

• Q2 2025: Stable PyTorch/scikit-learn APIs. 

• 2026+: Support for 24+ qubit photonic systems. 

Learn More: merlinquantum.ai 

WITTENSTEIN and Quandela underscore European innovative strength

Two leading technology companies from Germany and France are joining forces to help shape Europe’s future in quantum computing: attocube systems GmbH, a company of the WITTENSTEIN group and specialist in nanotechnology, and Quandela, a pioneer in photonic quantum computer technology. The companies have been working together on the development of the European quantum computer Lucy. Representatives of the owners, Management Board and senior management of the WITTENSTEIN group took advantage of a visit to Paris to meet with the Quandela team and assess the status of the joint project.

Lucy is no ordinary computer. It is based on light particles – known as photons – and belongs to a new generation of quantum computers that are opening up completely new possibilities in areas such as artificial intelligence, cyber security, and materials research. The quantum computer was commissioned by the European High Performance Computing Joint Undertaking (EuroHPC JU) following a competitive tender process won by the Quandela-attocube consortium.

The collaboration between Quandela and attocube demonstrates how European companies can work together to achieve technological excellence. While Quandela is developing the photonic quantum platform, attocube is supplying high-precision cryogenic systems—technology that generates the extremely low temperatures required for quantum processes.

The visit to France focused on technical progress and system integration. The participants discussed how quantum and classical computers can be combined even more effectively in the future—for example, for hybrid applications in AI or complex simulations.

“Lucy is more than a technical project – she is a symbol of European innovation,” said Dr. Bertram Hoffmann, CEO of WITTENSTEIN SE. Niccolo Somaschi, co-founder and CEO of Quandela, added: “Lucy stands for technological excellence and for the common goal of making Europe a world leader in quantum computing.”

Lucy is scheduled to go into operation later this year. It will be based at the French supercomputer center CEA TGCC, where it will serve as the cornerstone of a sovereign European quantum ecosystem.