Artificial Intelligence: Meaningless Marketing Term or Game-Changing Business Tool

Guest Author Feb 24 2017 | 03:45 PM | 5 Mins Read | Level - Intermediate | Read Mode

Seth Redmore Chief Marketing Officer, Lexalytics

Connect with Author

With over 20 years of combined experience in product management, marketing, text analytics and machine learning, Seth is currently the CMO of text analytics leader Lexalytics. Prior to this role, Seth served as Vice President of Product Management and Vice President of Marketing at Lexalytics. Seth has also held executive positions at both hardware and software companies, including co-founder of Netiverse (acquired by Cisco Systems). During his tenure at Cisco, Seth built Cisco's first internal text analytics solution for reputation management. Seth has a degree in Chemistry from Carnegie Mellon University.

Social Media Management vs Social Media Monitoring – What’s the Difference? 2 years ago

How AI can Optimize your Sales Funnel 3 years ago

Social Media Management vs Social Media Monitoring – What’s the Difference? 2 years ago

The widespread adoption of the terms “Machine Learning” and “Artificial Intelligence” point to their progression into an important component of the software industry. Seth Redmore, CMO at Lexalytics explains how brands can move past the marketing hype and make AI work for their business

AI. It’s a term that used to conjure up images from science fiction movies – the ones filled with machines that could think and act with human-like intelligence. In fact, for most of the last 40 or so years AI has been closely linked with things like the Turing test, that famous assessment for measuring a machine’s ability to exhibit intelligent behavior equivalent to or indistinguishable from that of a human.

Oh, how times have changed. That once-prestigious definition is long gone. AI is now nothing more than a marketing term used for any software application displaying even the most rudimentary intelligence. So, how did this happen? If you’re an engineer or data scientist the answer is both depressing and exhilarating. Put simply we have the success of Machine Learning/AI in applications like IBM Watson’s Jeopardy machine or Google’s new chatbot to blame. They’ve seen AI transformed from an almost unreachable goal into to a cute acronym designed to sell to the Fortune 500 and anyone with deep enough pockets.

Still, there’s a plus: the widespread adoption of the term points to the progression of Machine Learning into an important component of the software industry. Engineers and Data Scientists would be well advised to study and learn all they can about Machine Learning and AI - the technologies that will be driving most of the important new applications over the next decade.

Really, on the whole I’m quietly excited by what this means for the field. If Marketing has adopted the term, then the technology is becoming useful, and businesses are beginning to see the benefits of AI. But we still need to tread carefully. As exciting as this all is, the rule of caveat emptor applies – there are all sorts of companies and applications claiming to do AI when in reality most aren’t even close. Do your due diligence and keep your checkbook close – just because something says it does AI or claims to have the highest accuracy ever measured on the IMDB dataset doesn’t make it true.

Still, we can certainly accept that AI is making progress – and quickly. Chances are it will come up as something to be weighed by your organization. So, let’s take a look at how you can make AI work for you.

1. Don’t believe everything you hear

Claims relating to the IMDB are one of my biggest pet peeves, as they’re a meaningless accuracy measure. The IMDB set is popular because it’s free and has been scored by humans, making it easy to measure against. The problem? It’s a static set of content that doesn’t reflect the dynamic nature of the real world. You can train and tune against the IMDB forever, but real world doesn’t afford you that luxury. As a result, your AI models need to be able to grow and evolve as the world around them changes.

2. Don’t use Machine Learning/AI just to use it

Understand the needs of your business and make a proper business case for AI. Given how young the field is, make sure those needs are small and precise enough to yield good results if Machine Learning is employed. Machine Learning isn’t ready to solve giant, complex systems yet, but it’s ideal for highly specific point problems like:

Identifying a common set of complaints about a product, regardless of the language used to talk about the product.
Learning the jargon related to a particular line of business.
Identifying the side effects of a new treatment or drug.

All of these are point solutions – a solution to one particular isolated problem – and are ideal for Machine Learning/AI. On the other hand, if your project of choice is trying to build Westworld, you’re almost certain to fail.

3. Work with vendors who have been in the field for a while

Someone may been knowledgeable about AI, but if they haven’t deployed a Machine Learning/AI project in a commercial setting – or if they haven’t worked with someone else’s data, then you might want to rethink signing that contract.

It’s a brave new world out there. AI may be in its infancy, but I’m expecting some amazing advancements in the next few years. So do your homework and enjoy the enhancements this technology will bring to our lives via advances such as smart chatbots like the new Google Assistant or intelligent recommendation engines such as Amazon’s. These B2C examples are just the beginning, however. Before long similar AI will show up in B2B applications all over the enterprise. So, think about how and why you want to leverage Machine Learning, and you too may be able to make your job a whole lot easier – just like something out of a science fiction movie.