In the final installment in this series about federated AI, we wrap up the discussion around the potential benefits of federated AI in developing good models in healthcare. Remember that this extensive discussion of federated AI was preceded by our examination of the many obstacles that exist in healthcare in getting enough good data to build high-value models. And, our deep look into federated AI and learning as a possible solution is meant to start breaking through the conventional approaches to getting the data. For the foreseeable future, healthcare data will be siloed and the owners of it will be cautious in sharing it with those who are looking to build models. Continuing to go door to door to get data, hoping to patch enough of it together to create a good dataset that can live up to its promise in the real world is slow and painful process. Continuing with this approach will mean that the real adoption of AI in healthcare will take longer than necessary. So, we should start thinking outside the box to see how this critical bottleneck can be overcome.