The Top Artificial Intelligence Prediction for 2022: Composable AI

The composability precept is fascinating—and galvanizing organizations—for several reasons. It’s the key to nimbly adapting to the sometimes seismic shifts in business climates that unexpectedly arise. It’s also one of Gartner’s top 10 strategic predictions for 2022 and beyond.

But according to Indico Data CEO Tom Wilde, it’s something altogether else that could very well be of even more importance to firms today. It’s a means of competitive differentiation, and one of the most distinguishable available to enterprise users, at that.  

“All organizations, it doesn’t matter what industry you’re in, recognize that their unique ability to codify the work that they do is a competitive advantage,” Wilde explained. “That codification comes from the kind of investments they made in technology and the employee experience and customer experience.”

Investments in composable AI solutions enable the sort of codification Wilde referenced while allowing firms to build applications, workflows, and business processes with a modular approach that’s rapidly interchangeable to suit the particularities of any use case—or business condition—that arises.

The ability to support composable applications (touted by Gartner as its No. 5 top strategic technology trend for 2022) for AI is one of the reasons Indico Data has experienced impressive growth and development over the past 18 months, a fact reflected in the recent additions of CMO Jeff Thomas and CRO Bob More.

With a convincing confluence of elements for basing AI modules on (including intelligent software agents or bots, transfer learning, and Intelligent Process Automation), the company serves as an excellent case study for how far composable AI will go in the next 12 months. 

Competitive Differentiation

To paraphrase Gartner, composability is a mindset, technological resources, and operational capabilities to quickly adapt to business needs by focusing on modularity for business assets or applications. Composable applications, then, consist of modules that can be put together like building blocks to address the needs of a particular deployment. The many boons of this approach don’t just involve an inherent flexibility, adaptability, and creativity for building and implementing AI applications.

They also include a means of assembling applications in such a way that it distinguishes the experience of employees and customers from one organization to the next. Wilde articulated a use case in which “it doesn’t matter if five companies are all in commercial real estate. The way they codify the work they do in commercial real estate is differentiated and unique. They don’t just want an off-the-shelf software application that all their competitors can buy, because that’s not going to drive that competitive differentiation going forward.”

Composable AI

Composable solutions for AI deliver an extreme amount of competitive differentiation and flexibility for organizations deploying them. Indeed, depending on their specific use cases, no two composable AI solutions will be alike—particularly when relying on a platform as multifaceted as Indico Data for conquering the unstructured data inundating most firms. One can argue that this platform has always been composable in nature. It relies on a transfer learning approach with an assortment of deep learning models devised for specific verticals or use cases like object recognition, Natural Language Processing, and others.

The solution also utilizes bots (affectionately termed “docbots”, in some instances) to create action from models, like transferring unstructured text to specific structured data systems that need such information. The platform also plays in the process automation space with capabilities for entity extraction and IPA that are integral to streamlining workflows and completing them with maximum efficiency. These different elements empower firms to “create smart AI systems in a component approach bringing these modules together and feeding data into it,” Thomas noted. Consequently, organizations can mix and match these modules to compose their applications of choice to tame any unstructured data use case.  

User Experience, Too

The ultimate beneficiary of composable AI applications may be the enterprise users that can assemble these components in a way that’s not dissimilar to how they can assemble no code or low code applications. In this respect, composable AI creates an optimal user experience that, in many ways, simplifies AI for those who may not necessarily be technical subject matter experts in this branch of the data ecosystem.

“This notion of composability is really designed to describe to the enterprise customer what kind of acute fabric are you building and deploying at your company, but which results in more efficiency,” Wilde commented. “You know, a simpler way to complete a task rather than more complexity, which gets you the opposite.”

Contributor

Jelani Harper is an editorial consultant servicing the information technology market. He specializes in data-driven applications focused on semantic technologies, data governance, and analytics.

Opinions expressed by contributors are their own.

About Jelani Harper

Jelani Harper is an editorial consultant servicing the information technology market. He specializes in data-driven applications focused on semantic technologies, data governance, and analytics.

View all posts by Jelani Harper →