Trained business users will indeed find it easy to help themselves by creating simple automation content for their own use.However, creating anything stable, reusable, and at a higher scale requires a hard-core IT talent, which is most likely to reside within IT departments. IT is also home to other, often competing methodologies like business process management (BPM), integration-Platform-as-a-Service (iPaaS), traditional integration, and software development. It becomes critical to manage RPA within the same bag of tools to ensure they all happily coexist. Often labeled in IT circles as temporary and an unstable integration solution, RPA may not even find its way into the bag of tools, especially in the organizations with limited IT budgets. Even if a business manages to establish its own citizen RPA force, scaling up has to be a conscious management decision. Otherwise itf left to a number of random forces, it will eventually decide on the level of its adoption, both organization and market-wise.

Is Citizen Artificial Intelligence (AI)of Better Luck?

AI in an organization will mostly likely materialize as tens or hundreds of micro-rules, each providing an answer to a very narrow question of interest; for example: whether to grant a loan, what to say next in a chat conversationor whether to purchase or sell a stock. They are likely to serve as a horizontal service layer of intelligence to be orchestrated by a variety of workflow engines and custom business applications and eventually presented over various channels. The most popular of them will eventually find its way as a native part of business applications. As a horizontal reusable component that does not step on the toes of competing methodologies, AI has a much better chance of securing its place in the enterprise architecture. Additionally, the scarce hard-core data scientist resources vs huge number of narrow business problems, where ethical concerns are at stake, does not sound like a scalable proposition. This also contributes to the importance of citizen developers and data scientists as catalysts of business problems to be solved by machine learning.

Future of Work Vs. Future of Automation.

It is quite safe to say that the future of work will fill a spectrum between fully autonomous systems and intimate the cooperation of humans and machines. It is yet unclear though what kind of machine portfoliois likely to serve that spectrum. For now, AI appears to have its own clear role. RPA’s future may not be guaranteed yet. API-enabled business applications will eventually replace legacy ones, and iSaaS platforms already offer RPA-like capabilities on top of these APIs. Organizations with established BPM and integration practices may see RPA as a burden. There are already speculations of the RPA bubble due to inflated and unrealistic expectations, which is reflected in the multi-billion valuations of RPA companies. An eventual bubble explosion might stabilize the RPA market to three main areas. First, the organizations that prove their citizen RPA approach valuable. Second, organizations that do not want to bother with too many automation flavors. And third, specific use cases in areas like customer service and education, both human-to-machine and vice versa, which naturally calls for capturing, learning and replicating user actions throug, for instance, robotic desktop automation (RDA) and process mining.