Artificial Intelligence 101: Separating Hype from Reality

A lot of digital ink has been spilled about artificial intelligence (A.I.)’s potential impact on people and businesses. But let’s face it: much of the discussions about A.I. advancements are based on speculation. Organizational leaders must separate the hype from reality and take a hard look at A.I. technology before they can arrive at a viable strategy. Here, we’ll examine two areas where A.I. has made a significant impact, its limitations, and what to expect moving forward.
 
A quick Google search of artificial intelligence will bring up headlines like “How Artificial Intelligence Will Change Everything,” and “9 Ways Artificial Intelligence Will Affect Our Lives.” In other words, we’ve reached the top of the A.I. hype meter.
 
What is often overlooked are the technical and operational processes that must be put in place to take advantage of A.I. technology. Platforms, tools, and applications are only as useful as the data that’s provided. However, streamlining databases and records to make them usable is time-consuming and sometimes highly regulated. Furthermore, A.I. systems must “learn” how to perform duties, but the process of inputting data and training these machines can be complicated and onerous.

Supply Chain Innovation
Although A.I. technology includes significant challenges, that shouldn’t dissuade companies from exploring practical implementations. Case in point: global supply chains. It may not be as exciting as a self-learning robot, but A.I. technology has already made a significant impact on supply chain planning.
 
Behind many products and services is an often-complex supply chain. A medical device, for instance, may have thousands of parts that are sourced from various companies around the world. A few years ago, there was a scarce amount of data about parts manufacturers and their inventory. Part of the problem was that humans couldn’t keep up with large amounts of data, much less ensure that it was consistently updated.
 
Today, thanks to artificial intelligence and data analytics, data systems can process massive amounts of data to provide insights such as the ideal time to buy/not buy materials and determine which supplier offers the best price. Databases that are aligned with other sources of information, like weather and climate conditions, can also predict which suppliers will produce the best raw materials, based on their location. 
 
A.I. technology has helped companies shave 10 to 20 percent off supply chain costs, which for some manufacturers, could mean hundreds of millions of dollars in savings.

Smarter and Leaner IT Infrastructure
Another A.I. success story involves IT infrastructure. Over the past 10 years, the IT department of most companies has evolved from a few on-premise solutions and a handful of partners to dozens of cloud providers and hundreds of client partners or more. As a result, the IT department is responsible for managing and monitoring multiple networks handling vast amounts of data.
 
A.I. platforms can be trained to recognize different types of data issues and play a key role in helping employees prioritize issues and proactively handle problems before they escalate. Certain tasks can also be automated, allowing the staff to focus on solving difficult issues. A.I technology isn’t perfect—humans must still sift through false-positive errors—but it is rapidly improving, enabling employees to be more productive while reducing costs.      
 
Customer Service Obstacles
Although artificial intelligence shows a lot of potential in helping people work more efficiently and effectively, we are a long way from it fully replacing human workers. For example, A.I.’s limitations become clear in a customer service setting.
 
Last year, a number of brands such as Taco Bell, Domino’s, and 1-800-Flowers were early adopters of A.I-powered chatbots, but as Digiday reports, within a few months, many have scaled back these efforts. “Fashion retailer Everlane, which was one of the first Facebook Messenger partners, announced last week it would no longer use it, saying they would rather stick with email. For another conversational commerce pioneer, Spring, customers have found that the bot, based on Facebook Messenger, is hard to use and doesn’t have the level of personalization people expect.”

Most bots simply aren’t ready to handle the complexities of conversation and require human intervention. Or they are deployed without an understanding of the customer perspective. We do see some great potential uses of A.I. on the backend as a guide for associates who are speaking with a customer on the phone or chatting via text or SMS, or on social media. Linking A.I. with a knowledge management system can provide insights like the next best offer to give a customer or which question the customer is likely to ask next. Expect to see more cases in which A.I. is applied to collaborative tasks like this, instead of managing entire customer interactions.
 
A.I. offers tremendous possibilities to better serve customers and drive profits, but it’s important not to get caught up in the hype. Companies should continue to experiment with effective ways to integrate A.I. into customer service and other areas, but tread lightly and always give customers the option to connect with a human employee. After all, it’s unquestionable that A.I.’s influence is growing, but taking a realistic approach will enable companies to separate science fiction from reality. 


Like this post? Subscribe to our customer experience blog.

Also, check out the most recent issue of our monthly customer experience eNewsletter, Dialogue.