Predictions of machines being able to think like humans have been around for many years.  In 1958, Herbert Simon wrote, “…there are now in the world machines that think, that learn, and that create.  Moreover, their ability to do these things is going to increase rapidly until – in a visible future – the range of problems they can handle will be coextensive with the range to which the human mind has been applied.”1 Almost a decade later, he predicted that by 1985, “machines will be capable of doing any work that a man can do.”2 In 1967, Marvin Minsky, head of MIT’s artificial intelligence lab, stated that “within a generation the problem of creating ‘artificial intelligence’ will be substantially solved.”  He went on to add that “within 10 years computers won’t even keep us as pets.”3

Over the years AI researchers have often been overly optimistic.  Yet they have consistently struggled with two very basic limitations – insufficient memory and slow processing speeds.  These problems have kept AI grounded for years.4 So then why does it seem that artificial intelligence has suddenly sprung into the limelight?

Over the past 20 years hardware speed has improved by at least 100X, and key algorithms have improved 10X-100X.  These can combine to bring improvements of up to a million-fold in some applications, according to MIT’s Tomaso Poggio.5 And then there is the data providing fuel.  In 2017 an IBM study proclaimed: “Every day, we create 2.5 quintillion bytes of data.  To put that into perspective, 90 percent of the data in the world today has been created in the last two years alone – and with new devices, sensors and technologies emerging, the data growth rate will likely accelerate even more.”6 On top of all this we have global networks connecting millions of devices, mobile technologies growing our reach to every corner, cloud capabilities exploding storage and computing capacities, voice and image recognition expanding our inputs.  It’s easy to see why we are living in a world of accelerating change…and artificial intelligence applications are contributing to that acceleration.

High Expectations But…

These conditions have led to high expectations among executives.  A study by MIT Sloan Management Review in conjunction with The Boston Consulting Group7 found that almost 85% are looking for AI to help obtain or sustain competitive advantage.  In addition, 75% believe it will help their companies move into new businesses.

But even with these significant expectations, less than 40% of companies have an artificial intelligence strategy in place and only 19% are actively adopting AI into their products and processes .

What Stands In the Way?

Some companies cite legacy technology infrastructure as hindering AI progress, and many recognize that the cleanliness of their data is lacking.  Policies, regulations, and rights around data protection/privacy also pose some concern.  But most barriers are more related to people.  You will notice that many of these barriers run parallel to frequent digital transformation roadblocks.  That should be no real surprise since data-driven decision-making is a foundational element of digital transformation.  For example, PWC’s Digital IQ Survey8 found that only 20% of executives believe their company’s AI skills are highly developed.  With the growing technology spend outside IT, there may be fragmented AI experiments in various departments.  Such a siloed approach can be more costly but certainly restricts the big-picture view that can yield the greatest benefit.  This in turn causes more difficulty in building business cases to justify the investment resulting in failure to gain sufficient priority for the investment.  Change resistance and concern over potential job loss can also stymie AI progress as fear, uncertainty, and doubt prevail.

Significant Benefits Await

Artificial intelligence implementation promises a powerful boon to business.  An average 39% revenue increase by 2020 is expected by organizations that are currently using or plan to deploy AI technologies.  At the same time these companies anticipate reducing costs by 37%.9  Productivity gains will come from automating processes (including use of robots and autonomous vehicles) as well as using assisted/augmented intelligence to augment the existing labor force.  Consumer demand is expected to increase due to personalized and/or higher quality AI-enhanced products and services.  PWC predicts this could add as much as an eye-popping $15.7 trillion to the global economy in 2030, more than the combined output of China and India combined.10

AI applications are already at play in your everyday life.  Voice recognition apps like Alexa, Siri, and Cortana employ AI.  Facebook uses AI with image recognition to identify familiar faces from your contact list.  Ecommerce sites like Amazon use it to personalize recommendations based upon previous purchases or activities.  Google uses it to improve searches.  Google Maps suggests the fastest routes by analyzing traffic speed gathered from anonymous smartphone location data.  Financial companies, including PayPal, use AI algorithms to combat fraud.   Music services like Spotify and Pandora personalize their offerings through AI.11

It’s time to get your company onboard as well.


1Herbert Simon and Allen Newell, “Heuristic Problem Solving:  The Next Advance in Operations Research”, Operations Research, 6 (Jan-Feb 1958), 6.

2Top 10 Bad Tech Predictions –

3The Myth of Artificial Intelligence –

4A Brief History of Artificial Intelligence –

5What’s Driving the Machine Learning Explosion –

610 Key Marketing Trends for 2017 –

7Reshaping Business with Artificial Intelligence –

8PWC Global Digital IQ Survey:  Emerging technology insights –

9Amplifying Human Potential:  Towards Purposeful Artificial Intelligence –

10Sizing the Prize – What’s the real value of AI for your business and how can you capitalize –

1110 Real-World Examples of Machine Learning and AI [2018] –

About the Author:

Tag/s:Artificial Intelligence,