IBM is repackaging Watson, its artificial intelligence (AI) software that got famous for winning the game show Jeopardy. The new Watson AI and data platform aims to help companies integrate Watson into their business. IBM also unveiled new tools and services for data scientists who build Watson-based applications, making accessing and sharing extensive data sets easier.
The company said its Watson Data Platform provides the world’s fastest data ingestion engine and service that speeds up the process of building and deploying machine learning models. It’s available in both public and private cloud environments. It also offers a service that allows developers and data scientists to create and test data sets in the same environment where their application will run, allowing them to see the impact of changes to data as they make them.
A separate service enables organizations to tag and enforce their governance policies to prevent sensitive information from being accidentally shared or used outside of an approved use case. It’s an essential feature because as AI applications are built and deployed, they often need to be able to read from different data sets. It also helps companies comply with regulatory requirements.
Another new tool is called Data Catalog, which automatically tags and labels data so it can be incorporated into AI applications. It can also search and find data in a data lake or warehouse, where it might be stored in different formats. The third new tool is Watson Data Science Experience, which combines data science and AI tools, allowing teams to work more effectively on datasets.
One of the significant advantages of Watson is that it can sift through reams of documents and other information to identify patterns and provide insights. That can help businesses be more efficient, for instance, in the insurance industry, where a doctor could use Watson to pore through medical reports, research papers, and EMR data and suggest treatments that would be most effective for a patient.
But Watson’s high cost at the time made it a challenge for some companies to use it, and even today, it is expensive to deploy. It costs a million dollars or more annually, depending on the usage. As a result, it can be cheaper for companies to build their own AI software, which is what many are doing.
A new version of Watson is being released this week. It’s designed to be a more affordable way for enterprises to leverage the technology without investing tens of millions of dollars upfront. It’s also being integrated with other products from IBM and its partners. The new product includes analytics and data management capabilities and a conversational assistant that can answer questions in real-time. It’s being launched as part of a significant effort to get more customers on the IBM Cloud. Reuters notes that the move comes over a decade after IBM’s software called Watson got attention for winning the game show Jeopardy.
