How to prep your enterprise for the AI Revolution

Add bookmark

AI Artificial Intelligence

One of the most frequently discussed topics among business professionals is undoubtedly Artificial Intelligence (AI). Although the potential benefits of AI are well known, what preparation is necessary to make implementing this technology a truly competitive advantage for the enterprise?

AI is not a new term. It was first used in the summer of 1956 as a topic of discussion at the Dartmouth Conference. Even so, the term did not come about overnight but rather after hundreds of years of philosophical contemplation on the process of reasoning –dating as far back to the Ancient Greeks.

This new hype largely owns itself to the growth of Big Data, or structured and accumulated data, that now enables AI to be realistically used within business functions.

What exactly is AI?

AI is a branch of computer science that attempts to mimic human decision making. The term is often used interchangeably with Machine Learning, which is a sub-branch of AI that uses data to learn and make educated decisions.

SEE ALSO: The State of Robotic Process Automation and Artificial Intelligence in the Enterprise

Essentially what AI does is find patterns in data to predict business failures, successes and other courses of action.

Although many professionals see the apparent value of AI, there is still a disparity between the expectation of what AI can accomplish for their businesses and the amount of effort needed for it to achieve scale.

According to a study done by MIT Sloan Management Group and BCG, “16% of respondents strongly agreed that their organisation understands the costs of developing AI-based products and services. And almost the same percentage (17%) strongly disagreed that their organisation understands these costs. Similarly, while 19% of respondents strongly agreed that their organisation understands the data required to train AI algorithms, 16% strongly disagreed that their organisation has that understanding.”

Nearly as many organizations that understand the amount of resources required to successfully implement AI, also have little to no understanding of the data itself. In other words, there is a chasm between industry players who appreciate the work involved and those who do not. Moreover, the study revealed that those who have a greater understanding of AI have incorporated the tech extensively in processes and offerings.

There are three major factors, besides understanding, that predict successful AI initiatives: comprehensive data, well-established processes and internal AI professionals.

Ample and comprehensive data fuels AI

Big Data made AI possible for business use and it remains the key to making AI a success. AI reads data in order to find patterns in processes and situations and then it provides predictions based on this data for the next course of action. Without an exorbitant amount of readable data, AI cannot provide accurate, predictive results.

Beside being abundant, this data must also be comprehensive, providing both records of business successes and failures. According to Jacob Spoelstra, Director of Data Science at Microsoft, “A mistake we often see is that organisations don’t have the historical data required for the algorithms to extract patterns for robust predictions. For example, they’ll bring us in to build a predictive maintenance solution for them, and then we’ll find out that there are very few, if any, recorded failures. They expect AI to predict when there will be a failure, even though there are no examples to learn from.”

Considering this, the first major and probably most difficult step in preparing for AI implementation is cleaning up and collecting all the company’s data. But as many Business Intelligence analysts will account to, there is no real end to investing in the company’s data pool. This action of preparing data for AI is better viewed as a long term process.

AI replaces steps in business processes, not people

AI streamlines current business processes by taking over certain steps along a complex workflow. This means that enterprises must be able to look at their well-structured processes and determine wherein these processes AI can be most useful. Without well-established processes, implementing AI is likely to add more complexity to an already complex workflow.

Often the steps AI takes over are repetitive and operational. AI rarely replaces entire workflows and the idea that AI will make many jobs obsolete is only partly true. AI is expected to free the human workforce from doing many monotonous tasks but not necessarily leave thousands or millions without work.

It is probable that more skilled labor will be needed with the expansion of AI and that the current workforce will need to learn new skills. However, businesses that implement AI are expected to grow and expand, leading to overall job growth.

This was the case with the popularisation of ATMs in the U.S. from the 1970s and onward. Initially, the amount of tellers decreased but for over a span of 40 years, ATMs allowed banks to save money on operations. In turn, this led to exponential growth in the number of branches and jobs until 2010 when advances in ATM capabilities led to another drop in the demand for tellers.

In the short term, what many enterprises (even with well-established processes) don’t consider thoroughly is creating a fluid pass-off to and from human and machine workers.

Julie Shah, an Associate Professor of Aeronautics at MIT, says, “What people don’t talk about is the integration problem. Even if you can develop the system to do very focused, individual tasks for what people are doing today, as long as you can’t entirely remove the person from the process, you have a new problem that arises — which is coordinating the work of, or even communication between, people and these AI systems. And that interaction problem is still a very difficult problem for us, and it’s currently unsolved.”

One way to combat this integration problem is by selecting an agile orchestration layer that will help the transfer of work and information to be more fluid. Some agile BPMs and other cloud-based softwares can ensure this layer is both business user friendly and AI compatible.

In-house AI professionals are worth the investment

AI can’t simply be implemented and ‘run’ like app software. AI algorithms need to be trained. According to the MIT & BCG research “Training AI algorithms involves a variety of skills, including understanding how to build algorithms, how to collect and integrate the relevant data for training purposes, and how to supervise the training of the algorithm.”

The algorithms not only need collected data to run, but they also need to be shaped by trained professionals who know how to properly feed, write and monitor these algorithms for long-term business performance and scale.

Digital transformation is a necessity for AI

In conclusion, preparing for AI takes active steps toward collecting data, developing well-structured processes, and hiring professionals that will dedicate toward long-term success. These are the digital foundations for AI adoption. Without this base, AI is an unrealistic step and could be a waste of business resources.

Like this content? You may also enjoy 7 Ways the Rise of Digital Technology Impacts Process Excellence.

Or watch the latest expert insight on PEX Network's video channel: