Before COVID-19, the majority of companies weren’t turning their data into gold with artificial intelligence. Today, most still have advanced AI in their imaginations more often than their information technology departments. But some did hit a few of their AI targets in one short year, and they have lessons for the more sluggish among us.

Right now, live AI use cases include: customer care, employee experience, AIOps, regulatory compliance and financial planning and budgeting, according to Rob Thomas (pictured), senior vice president of software, cloud and data platform at IBM.

“We have 100 customers that went live with one of those two use cases in the first 35 days of the pandemic,” Thomas said. “I think without the pandemic, for those 100, it might have taken three years.”

This provides encouragement in the face of relatively dismal statistics on big data and AI deployments. For instance, a report from Seagate Technology PLC from last June revealed that said 68% of data in organizations goes unused. Looking at IBM’s customers, it’s possible to conclude that the problem isn’t a lack of effective technology. The tech was already there; the straits imposed by the pandemic forced them to finally put it to use.

“Sometimes it just takes the initiative to get started and to do something. And all those companies have really benefited from this, so it’s great to see,” Thomas said. 

Thomas spoke with John Furrier, host of theCUBE, SiliconANGLE Media’s livestreaming studio, during IBM Think. They discussed what holds companies back from deploying AI, as well as technological and cultural solutions. (* Disclosure below.)

Automation’s the link between data and AI

So why didn’t everyone else advance on their AI roadmaps in 2020? Motivation, top-down strategies and cultural shifts are needed at some companies. Others have data limitations built into their architecture that they must remedy, according to Thomas. Architecture for data intelligence must include automation, for example. The more automation companies can build into the data systems that feed analytics and AI, the better.

“Probably the biggest bet IBM has made in the last year is in automation — explicitly in Watson AIOps,” Thomas said.

IBM acquired automated-observability platform Instana last year and also announced it would acquire Turbonomic AI app-resource management. Also, IBM’s AutoSQL automates the creation of SQL queries in large, distributed data warehouses.

“It’s never been done before. Now we’re doing it,” Thomas said. “At this point, we’re the only company that has all the pieces for automating how you run your IT systems. That’s what I mean when I say AIOps.” 

There are also simpler software tools and add-ons on the market to help companies quickly put AI into practice in some areas. Ready-to-use business software, like assistants with natural language processing, can enable regular business users to complete tasks faster. IBM’s Maximo Mobile has Watson AI built in so field technicians can use supported devices for improved troubleshooting, repairs and the like.

With AI tech continually improving and adoption picking up, companies can hardly afford to remain lazy. “AI won’t replace managers, but managers that use AI will replace the managers that do not,” Thomas said. 

Thomas’ final piece of advice is to pick from the five proven AI use cases mentioned at the start of this article. “Pick one of these and get started, then pick the second, and then pick the third,” he advised. “If you are not doing all five of these 12, 18, 24 months from now, you are going to be behind.” 

Here’s the complete video interview, part of SiliconANGLE’s and theCUBE’s coverage of IBM Think. (* Disclosure: TheCUBE is a paid media partner for IBM Think. Neither IBM, the sponsor for theCUBE’s event coverage, nor other sponsors have editorial control over content on theCUBE or SiliconANGLE.)

Photo: SiliconANGLE