The activity develops and validates an AI-enhanced GNSS integrity monitoring layer for  safety-critical positioning applications. The solution combines high-accuracy GNSS  correction services, RTK positioning, protection-level estimation and machine-learning based anomaly detection to assess whether a position solution can be trusted in real time.  By monitoring correction quality, receiver behaviour, satellite geometry, residuals and  operational context, the system generates integrity indicators and safety-relevant status  outputs such as OK, degraded or alert conditions. The work targets demanding use cases  including autonomous and mixed-traffic vehicle operations, robotics, industrial sites and  critical infrastructure monitoring, where positioning errors must be detected before they  create operational risk. The activity will deliver a validated prototype architecture, field-test  evidence and a roadmap toward operational deployment.