Edge AI
AI processing performed on local devices — smartphones, IoT sensors, cameras, embedded systems, vehicles — rather than in the cloud. Edge AI keeps data local, reduces latency, and enables real-time decisions, but introduces governance challenges around update management, monitoring, and hardware constraints.
Why It Matters
Edge AI trades centralized control for speed and privacy. Governance teams must address how to update models deployed on thousands of devices, how to monitor for drift without centralized logging, and how to ensure consistent behavior across heterogeneous hardware.
Example
A fleet of autonomous delivery robots runs computer vision AI on-device for real-time navigation. The governance challenge: how to push model updates safely to 500 robots in the field, monitor for performance degradation without transmitting video back to the cloud, and handle incidents when a robot makes a bad decision offline.
Think of it like...
Edge AI is like franchising — you distribute your operations (AI models) to many locations (devices), gaining speed and reach, but you lose the centralized control that made quality assurance straightforward.
Related Terms
Computer Vision
A field of AI that trains computers to interpret and understand visual information from the world — images, videos, and real-time camera feeds. It enables machines to 'see' and make decisions based on what they see.
Model Drift
The gradual degradation of a model's predictive performance over time as the real-world environment changes. Model drift can be caused by data drift, concept drift, or both.
Post-Market Monitoring
Ongoing surveillance of an AI system's performance, safety, and compliance after it has been deployed to production. Required under the EU AI Act for high-risk systems, post-market monitoring ensures that AI systems continue to meet their intended specifications as real-world conditions change.