blobbotronic - stock.adobe.com
Most people don't understand what AI really is, according to Rob Thomas, general manager of IBM data and Watson AI.
"When they hear AI, they think of autonomous vehicles" or similar products, the 20-year veteran of IBM said. "We've lost touch with the notion that AI is about making better predictions."
That's partly why AI adoption in the enterprise is low now, Thomas said. Other factors come into play too.
For one, enterprises struggle with collecting data and putting it into usable form. That's important, as "your AI is only as good as your data," Thomas said.
According to Thomas, IBM has made efforts to tackle each of these impediments. The tech vendor has tools for structuring data and automating some of the day-to-day tasks data scientists would normally carry out, as well as new tools to help explain the outputs of AI models.
Rob ThomasGeneral manager, IBM data and Watson AI
IBM's Data Science Elite teams help develop use cases for clients and then mentor users in developing and deploying analytics and AI technologies.
Finding potential applications for clients is important, Thomas said. It brings the technologies to life and makes them more tangible for people.
"People are genuinely interested in and inspired in the way they can use AI," he said.
Financial potential for AI
AI promises huge financial rewards, Thomas said, noting that McKinsey has estimated that by 2030, AI will have generated $13 trillion in new global economic activity.
The financial picture also bears promise for enterprises looking to adopt AI.
"AI is actually going to create jobs. It's going to make people more productive," Thomas said.
"AI done right is about automating the things you don't want to do to create more business opportunity to do more things that you do want do," he continued. "If we can create a culture that's data-centric … then adoption will really take off."