In a recent press release statement, last week, IBM, an Armonk, New York-based multinational information technology (IT) company, announced the general availability of two of its Watson machine-intelligence services as standalone applications on the public or private cloud. The company will be offering these Watson AI services atop its IBM Cloud Private for Data, a combined analytics and data governance platform that supports Kubernetes deployment.
According to Ruchir Puri, CTO and chief architect for IBM Watson, the company working on customer’s demand for machine learning solutions for multi-cloud or hybrid cloud environment, where the customer data already resides, thought of this strategy. Adding further he said that instead of trying to move the data to a single cloud and create something called a lock-in in this open compute-environment-driven world, the company thought of making AI available to the data. Concluding he said that, the concept is similar to how Hadoop and other mass data-processing systems function, they work on data in place, rather than moving the data to the processing.
As of now, only two of the IBM flagship services namely, “Watson Assistant” and “Watson OpenScale”, will be available to customers as standalone applications. Watson Assistant can be used to develop “conversational interfaces” such as chatbots; & Watson OpenScale can be used to provide “automated neural network design and deployment,” i.e. training, deploying, and overseeing enterprise machine learning models and neural networks.