This presentation will introduce the concept of “hybrid AI” and discuss how it can help manufacturers more cost-effectively deploy advanced machine learning skills in existing applications. One of the key challenges for manufacturers evaluating AI is cost concerns related to replacing existing infrastructure and proven end-user processes. In comparison, a hybrid approach integrates AI with existing vision algorithms, infrastructure, and end-user processes to help reduce deployment costs and complexity. This session will highlight how machine learning differs from traditional machine vision and the advantages and drawbacks of the two approaches. In particular, we will highlight on how a hybrid approach combines the best of traditional vision algorithm and machine learning capabilities reduce inspection errors and costs. We will close the session with a real-world case study of a global consumer goods manufacturer now deploying hybrid AI in a consumer good inspection applications, and discuss some of their concerns, challenges and benefits as they have migrating to a hybrid inspection approach.