T-RIZE Joins Flower Pilot Program to Advance Privacy-Preserving AI

T-RIZE, a Montreal-based company, has made a significant stride in the realm of privacy-preserving artificial intelligence by joining the Flower Pilot Program. This three-month initiative aims to advance federated learning through practical applications, marking a pivotal moment in redefining privacy standards in workplace AI.

The company will develop and deploy a production-ready blueprint that enables enterprises to train AI models on dispersed datasets without compromising critical data. This blueprint, set to be released at the conclusion of the program, will demonstrate how to fine-tune pre-trained transformer models on various datasets, including rental records, underwriting data, and identity verification forms. The process will utilize Flower’s federated AI framework in conjunction with Rizemind, T-RIZE’s open-source blockchain-integrated AI library.

This strategic partnership positions T-RIZE at the forefront of an emerging industry focused on privacy-preserving AI, supported by cryptographic and decentralized infrastructure. The roadmap will equip institutions with the necessary tools to meet the growing demand for AI systems that are secure, auditable, and compliant with regulatory requirements.

Madani Boukalba, CEO of T-RIZE Group, emphasized the importance of this development, stating, “Enterprises want the benefits of AI, but they also need guarantees that their data is protected and their processes are verifiable. We’ve built Rizemind to address that challenge head-on by embedding privacy, traceability, and economic incentives directly into the learning process.”

Traditionally, AI development involved centralizing massive datasets, raising concerns about data leakage, vendor lock-in, and noncompliance with privacy legislation such as GDPR, HIPAA, and Canada’s Bill C-27. Federated learning offers a compelling alternative by allowing AI models to train across decentralized data environments while keeping the raw data intact.

T-RIZE enhances this approach with Rizemind, which integrates blockchain-based auditability, incentive layers, and on-chain coordination into federated learning processes. This integration records every model contribution on the Rizenet blockchain, with incentives delivered in the $RIZE token—a utility asset used for compute credits, participation prizes, and performance validation.

The blueprint includes an open-source GitHub repository with transformer model code, Docker images for quick deployment, checklists and dashboards for schema validation, performance monitoring, and data alignment, as well as built-in encryption, access control, and network isolation to meet enterprise-grade security standards.

The Flower framework, utilized by prominent organizations, is recognized as the industry standard for federated AI research and implementation. Its open environment and modular architecture make it ideal for working with blockchain-based solutions like Rizemind. The collaboration between Flower and T-RIZE aims to bridge the gap between machine learning efficiency and regulatory-grade transparency, enabling organizations to train stronger models without relinquishing data ownership. This partnership also creates a tokenized framework that aligns incentives for data sources, model contributors, and validators.

Dimitris Stripelis, a representative for the Flower Pilot Program, highlighted the significance of this partnership, stating, “T-RIZE brings something powerful to the table: a verifiable, decentralized way to collaborate on AI. This partnership is about demonstrating how AI can be both private and accountable at scale.”

Looking ahead, T-RIZE’s roadmap includes further enhancements to Rizemind, such as zero-knowledge machine learning (zkML) for on-chain verification of model correctness, multi-party computing (MPC) to reduce data risk, and dynamic noise addition to protect privacy in sensitive contexts. This long-term vision underscores T-RIZE’s commitment to advancing enterprise AI while prioritizing data privacy and compliance.