Machine Learning Prague 2026, the 11th international conference on Machine Learning and AI applications, brought together over 1,000 attendees, 45 speakers, and 10 workshops across three days at O2 Universum in Prague. This year, CERTAIN had the opportunity to present at the conference’s poster session, showcasing our work on ethical and regulatory AI certification to an audience of researchers and practitioners from across Europe and beyond.
CERTAIN at the Poster Session
The poster session, held during the lunch break on May 5th, featured a diverse range of academic and industry work. Fabian Kovac from the University of Applied Sciences St. Pölten presented the CERTAIN poster titled “A Framework for Ethical and Regulatory AI Certification with Ontology-Driven Transparency”, highlighting the project’s core contributions: the AIDOC-AP ontology, the extended NeOn-ACE pipeline, and the Semantic MLOps infrastructure for automated EU AI Act compliance.
What Sparked the Conversation
The reception was genuinely encouraging. Attendees were particularly drawn to the methodology (how CERTAIN systematically approaches the translation of dense legal requirements (EU AI Act Annex IV) into machine-readable ontological constructs, and how the NeOn-ACE pipeline introduces iterative LLM-assisted coverage evaluation and semantic alignment as novel engineering steps).
The EU-funded nature of the project sparked natural curiosity, and the consortium’s multi-partner structure (combining academic expertise from USTP with real-world energy community deployments at EMPOWER and data space integration from Digital for Planet) resonated well with an audience used to seeing single-team academic work.
Perhaps most notably, people responded to the practicality of what CERTAIN is building. Given the complexity of the underlying domain, visitors were pleasantly surprised by how approachable the first tool versions are to set up and use. The general sentiment was that the project addresses a real and growing pain point in the European AI landscape, and that people will be following its further development closely.
Looking Ahead
CERTAIN is still in active development, with full end-to-end integration of the Semantic MLOps Engine, AIDOC-AP ontology, and RegOps compliance assessment targeted for later this year. If you’re building or auditing high-risk AI systems under the EU AI Act and want to explore what automated, ontology-driven compliance could look like for your use case, we’d love to hear from you.
🔗 Website: https://certain-project.eu/
🔗 Ontology: https://w3id.org/aidoc-ap
🔗 GitHub: https://github.com/CERTAIN-Project
