As the initial hype for artificial intelligence fades, enterprises are getting down to the nitty gritty of how to actually deploy AI in the hybrid cloud landscape.

Optionality and a multicloud infrastructure are key for implementing AI into businesses, according to Hillery Hunter (pictured), chief technology officer of infrastructure and general manager of innovation at IBM Corp.

“Ensemble AI and combining technologies has been a core focus of us and what we’ve been doing in our program for enabling integration of AI on the mainframe systems,” she said. “Using a model that is small and efficient for the targeted purpose can help an enterprise get to that value stream generation … much more quickly. Providing that optionality, to combine together the best of both worlds, is a way that we’re seeing a lot of enterprises have success.”

Hunter spoke with theCUBE Research’s David Linthicum at theCUBE on the Ground – IBM Analyst Forum event, during an exclusive broadcast on theCUBE, SiliconANGLE Media’s livestreaming studio. They discussed AI trends within the industry and IBM’s strategy for giving companies greater flexibility. (* Disclosure below.)

Everyone has bought into the hybrid cloud landscape

Getting a return on investment for AI is not about finding one, perfect solution, but instead creating an adaptable, hybrid cloud system, according to Hunter, who estimates that over 80% of chief information officers use multiple clouds.

“Everyone is trying to get ROI out of AI, having now shown its ability to do something useful for the enterprise,” she said. “Part of getting ROI is getting the cost under control and we firmly believe that that means looking at AI as a conversation in its entirety across the full hybrid cloud landscape. There’s this optionality [with] hybrid cloud and hybrid cloud by design. Wherever your data is your customers are, you want to have the optionality to match the AI to those things so that you can get the cost efficiency to be as good as possible.”

IBM is focused on two strategies: creating smaller models that “punch above their weight” or perform multiple functions and adapting models for customized purposes. These types of AI services allow for a greater return on investment for IBM’s customers, Hunter believes.

“You can use the Granite model in public cloud, from an API perspective, through a service, you can build around it, you can get an appliance on-premises to use our models as well, or you can just consume it as software and run it on hardware you’ve already got,” she said. “That’s the flexibility. Start with software and appliance, a cloud service, whatever you need to get to a cost-optimized outcome — we’re trying to meet enterprises where they’re at.”

Here’s the complete video interview, part of SiliconANGLE’s and theCUBE Research’s coverage of at theCUBE on the Ground – IBM Analyst Forum event: 

Watch the full event video below:

(* Disclosure: TheCUBE is a paid media partner for the IBM Analyst Forum event. Neither IBM Corp., the sponsor of theCUBE’s event coverage, nor other sponsors have editorial control over content on theCUBE or SiliconANGLE.)

Photo: SiliconANGLE