The Future of the Digital Worker: Demystifying Intelligent Process Automation

By: Colin Redbond, Blue Prism's Head of Technology Strategy and Architecture

It is well known that Blue Prism invented RPA, but in this brave new world of Artificial Intelligence (AI), we are often asked, “is Blue Prism also Intelligent RPA?”

When Blue Prism coined the term “RPA” back in 2012, it applied to a new breed of enterprise scale, dynamic automation that we had developed by working closely with our clients. One that was able to reproduce the exact actions of a human operator; using the same interfaces, reproducing the same complex decisions, “seeing” what the human operators performing that task manually were seeing, adapting to variations in performance, environmental factors and changing data.

By the very nature of the Oxford Dictionary definition, Blue Prism would already be classified as AI, whereas other tools  based on more primitive “macro recording” technologies or limited by low level hard coded interactions would be unable to achieve the same level of sophistication.

In 2016, Blue Prism was proud to receive an award for best enterprise implementation of AI at the AIONICS awards, further underpinning our position as an Intelligent RPA platform.

So, to answer the question “is Blue Prism Intelligent RPA”?

It sure is! In fact, you could argue that we invented it, but no technology stands still, and we continue to develop our strengths, in conjunction with those of our broad partner ecosystem.

We also recognize that the definition of AI is evolutionary. This can be attributed to what is sometimes referred to as the AI Effect, where technology moves at such a pace that advances that could previously be categorized as AI will no longer be seen as such, as they become more mainstream.

Indeed, some even argue that “pre-packaged” Cognitive Services capabilities such as NLP and Computer Vision should no longer be classed as AI. In our view, the models which underpin these capabilities are highly complex, so they still qualify as AI. Whatever your view, one thing is for sure, we (and other vendors in the RPA market) have a responsibility to be consistent and honest with our messages, because, as the concept of “Intelligent Process Automation” moves up the hype curve, there will be a temptation to redefine the space with hyperbole.

At Blue Prism, we take a more pragmatic approach.  If we step back for a moment, there are some inherent dangers with seeking to define a “new breed” of RPA, as opposed to considering this as an evolution of the technology:

  • There is a tendency to overlook the real reason why you first began to look for an automation solution – business benefit.  When I’m asked about what intelligence we provide in our platform, or what our AI strategy is, my first response is to ask, “what is the business problem that you are trying to solve?”. In many cases, this is met with a slightly uncomfortable silence. One thing that is consistently proven when looking at successful RPA projects, forget about whether you class them as using weak, strong, narrow, wide, or any other level of AI, is that you need to have a strong execution platform before considering how you might “get up the value chain” of intelligent automation. A question sometimes asked of our partners is “shouldn’t I just skip RPA and go straight to AI?” Well, imagine if you were comprised only of a brain, with no eyes, arms and legs. A strong RPA execution platform and COE is essential as a pre-requisite to any form of Intelligent RPA.
  • Due to the “AI Effect”, it’s inevitable that the goalposts will be constantly moving. Should we define a new category of automation, which is based on a set of “check boxes” that may well be out of date and not considered as true AI in 6 or 12 months’ time, or should we instead focus on the true transformational value of an RPA product, it’s abilities and the business outcomes that it delivers? Inevitably, as AI technology matures, some of the abilities used by the Digital Worker will also include higher levels of Cognition and Intelligence, but these should be assessed holistically as one of a set of attributes you need in an RPA product to solve your business use cases.
  • There is a constant pressure for vendors to jump on the bandwagon and tout their AI capabilities, when the stable execution platform to underpin that capability is missing. Even worse, in some cases the quality or maturity of some capabilities being used is sub-par and will lead to disappointment or more failed projects. There is a reason why Blue Prism has a 100 percent renewal rate – we stand behind every claim that we make and we’re able to deliver an unmatched level of ROI. The recent Forrester Total Economic Impact (TEI) study highlights an ROI of 229 percent, with $49 million saved and payback in just 15 short months. This is virtually unprecedented in any technology field.

"The Prism" is a new series of blogs, where we will be exploring some of the key fields of Artificial Intelligence and how these abilities relate to the future of the Digital Workforce. These will include Natural Language Processing, Machine Learning and Computer Vision, amongst others. We will also look at the benefits of leveraging an Open API Economy and provide real, tangible business use cases of technology within the Blue Prism Intelligent RPA ecosystem.

We will explore how a strong RPA foundation built on the strengths of a scalable, extensible platform can ultimately become the gateway to understanding and accessing emerging intelligent technologies.

I hope you will join us for the journey.