# Bunq cuts false positives by 63% using AI. Fully compliant.

> See how Bunq cut false positives by 63%, sped up case handling by 56%, and reduced AI maintenance costs while staying compliant.

**Source:** https://deeploy.ai/success-story/bunq-3/

**About this file:** Machine-readable markdown version of the page above, for LLMs and AI assistants. Regenerated automatically whenever the page is updated.

See how Bunq replaced rule-based monitoring with AI to detect more criminal behaviour, speed up case handling, and stay fully compliant with the EU AI Act.

- **63%** Fewer false positives
- **56%** Faster case handling
- **30%** Lower AI maintenance costs
- **100%** Compliance ready

## About the company

**Bunq** is one of Europe’s fastest-growing challenger banks, offering millions of customers a **transparent, tech-driven, and customer-first banking experience**. To strengthen its defences against financial crime, Bunq turned to **AI-powered transaction monitoring** to replace outdated, rule-based systems.

### The challenge

- ✗ High False Positives: Traditional rule-based monitoring systems flag too many legitimate transactions, overwhelming compliance teams with unnecessary manual reviews.
- ✗ Lack of Trust in AI: Investigators struggle to rely on AI outputs, as decisions are hard to interpret and governance is limited.
- ✗ Team Misalignment: Development and compliance teams struggle to coordinate, creating inefficiencies and raising the risk of fines or reputational damage.
- ✗ Slow Model Updates: Models are difficult to update quickly, which reduces effectiveness and increases manual investigations.

## Adopting Deeploy

Bunq adopts Deeploy as the governance layer for AI-driven transaction monitoring. From one **centralised platform**, **all models are deployed and monitored**. Built-in explainability shows investigators why transactions are flagged, restoring trust in AI.

Compliance officers can **review and refine AI decisions** in real time, while Deeploy automatically creates audit trails and documentation ensuring every model stays aligned with the EU AI Act.

By connecting development and compliance teams, Bunq **improves models continuously** while ensuring transparency and strong **regulatory oversight.**

## Less false positives, lower AI maintenance costs

With Deeploy, Bunq transforms its transaction monitoring process. **False positives drop by 63%,** reducing unnecessary manual investigations. **Case handling becomes 56% faster**, thanks to explainability and investigator-friendly tools. **AI maintenance costs fall by 30%** as governance and feedback loops streamline operations. Most importantly, compliance is built in every decision is documented, auditable, and aligned with regulatory standards and EU AI Act.

> "Having effective human oversight and the ability to control the risks of AI within our organisation delivers clear insights and the possibility to train, monitor, and provide feedback to our ML models. From Data Scientists to Customer Support, everyone has the ability to understand the model outcomes thanks to Deeploy’s platform."
>
> — Ali el Hassouni, Head of Data, bunq

## All with governance & compliance built-in

For Bunq, Deeploy is not just a technical layer it **acts as a compliance safeguard.** By ensuring transparency, observability, and alignment between development and compliance, Deeploy enables Bunq to **innovate responsibly while protecting customers and meeting regulatory requirements.**

## See how you can also cut false positives by 63% with full compliance

**Transform AI monitoring with Deeploy.**

---

**Get in touch:** You can book a demo at https://deeploy.ai/book-demo/ or reach the team via https://deeploy.ai/contact/.