Federated AI Project TRUSTAM Targets AM Quality Assurance
Swedish software innovator Interspectral and a consortium of industrial and technology partners have secured funding from Swedish innovation agency Vinnova for TRUSTAM (Trusted Federated Intelligence for Additive Manufacturing), a groundbreaking initiative focused on advancing secure AI-driven quality assurance for additive manufacturing.
The project, which brings together Interspectral, Saab, AMEXCI, and Scaleout Systems, aims to solve one of the most pressing challenges in industrial AM adoption: how to scale intelligent process monitoring and quality assurance across multiple production sites without compromising sensitive manufacturing data or intellectual property.
The TRUSTAM project has been awarded SEK 7.4 million (€679,000) in funding through Vinnova’s “Industrial Applied AI by Advanced Digitalization 2026” programme and will run from April 2026 through April 2028.
At the core of the initiative is the use of federated learning – an emerging AI framework that enables machine learning models to be trained collaboratively across geographically distributed production environments while keeping raw process data localized and secure. Instead of transferring sensitive manufacturing data between facilities, only model updates are exchanged, allowing organizations to benefit from collective intelligence while maintaining strict confidentiality.
For sectors such as aerospace, defence, energy, and other safety-critical industries, this represents a major technological step forward. These industries often operate under stringent security and regulatory constraints that limit data sharing, making centralized AI training difficult or impossible.
According to Isabelle Hachette, the project directly addresses challenges increasingly faced by AM manufacturers as production scales globally.
“How do you scale AI-driven quality assurance across multiple production sites and different machine environments without ever compromising data security or IP ownership?” Hachette stated. “That question demands a collaborative answer, and this consortium is uniquely positioned to deliver it.”
Within the consortium, Interspectral will play a central technical role, leading development of the local AI models that operate directly on-site within manufacturing environments. These AI systems will learn from machine-specific process data and adapt to unique materials, machine configurations, and production conditions.
The project also builds upon Interspectral’s existing AM Explorer platform, which has gained traction across the metal AM sector for integrating and visualizing complex manufacturing datasets from in-situ monitoring systems, simulations, and post-build inspections. According to the company, AM Explorer is already deployed across a significant portion of the metal AM machine landscape and used by customers including GKN Aerospace and Volum-E.
TRUSTAM’s objectives extend beyond software development. The consortium plans to establish validated frameworks for secure cross-site AI collaboration and demonstrate the technology in live aerospace and defence production environments. The expected outcomes include on-premise AI models tailored to specific machine conditions, multimodal process analysis capabilities, and seamless workflows connecting monitoring, analysis, and manufacturing decision-making.
The project also reflects a broader strategic trend within advanced manufacturing: the convergence of AI, digital twins, and industrial data infrastructure to enable real-time process qualification and autonomous quality assurance.
Over the last several years, Interspectral has steadily expanded its role in this transformation through multiple Vinnova-supported initiatives. Earlier projects such as ‘AM Intelligence’ focused on automated defect detection and AI-assisted quality assurance in metal additive manufacturing, laying much of the groundwork for the federated AI approach now being pursued in TRUSTAM.
The company has also participated in major industrial collaborations including RES2AM and ESCROW, projects designed to improve sustainability, resilience, and process optimization in large-scale metal additive manufacturing using AI-powered monitoring and analytics.
As additive manufacturing continues transitioning from prototyping to full-scale production, particularly in regulated industries, the ability to prove quality in real time remains a critical bottleneck. Federated AI approaches such as TRUSTAM may provide a pathway toward scalable, secure, and certifiable AM production ecosystems.
The TRUSTAM consortium is expected to complete demonstrator activities by 2028, with results disseminated to the broader additive manufacturing community.
For more information: www.interspectral.com
