Five Things I Learned from MIT’s “AI: Implications for Business Strategy” Certification Program

Here’s the thing; as marketers, you need to be pretty aware of the various ways AI is impacting our lives. A few prime examples of AI in action include social media and newsfeed algorithms, social listening algorithms, and natural language processing bots and assistants. AI is officially 2018’s hyped technology, and chatter about its proliferation has not yet reached its apex. In fact, as our CEO, Barri Rafferty, noted, one of the primary focuses of the 2018 World Economic Forum was the impact AI is starting to have on our society.

With all of this in mind, I decided to seek out some subject matter knowledge through a certification program with the MIT Sloan School of Management called, Artificial Intelligence: Implications for Business Strategy. 

Here are some key lessons from the program…

AI isn’t going to destroy us.
The good news is that we’re nowhere near the dystopian futures that people have prophesied. Granted, there have been incredible advances, but we’re still pretty far away from living through a Hollywood, or Netflix, drama.

I’m an optimistic person by design and was pretty convinced after the MIT program that AI will not destroy our lives. The best thing us non-AI engineers and technologists can do is to stay up to speed with the research around the fields that will be most impacted, and make personal decisions on the best way to mitigate any issues that might arise. Or for that matter, if you’re a marketer, continue to learn new ways in which you can leverage it for clients.

Artificial Intelligence is actually #oldnews.
Artificial intelligence isn’t a new concept by any means. AI has been studied and developed since at least as early as 1950 and Alan Turing’s seminal work on artificial intelligence was written more than 60 years ago. His paper “Computing Machinery and Intelligence” opened with, “Can machines think?” The paper described what is now known as the Turing test, which helps humans evaluate a machine’s ability to show intelligent behavior that is equal to, or indistinguishable, from that of a human performing the same task.

Turing’s work almost inadvertently wound up putting a more philosophical definition behind “intelligence,” framing it in a way that made it conceivable for computers to convey intellect. It was from that point on that scientists began using machines to develop human-like mathematical logic that could perform tasks like solving puzzles. MIT’s AI Lab, then known as the AI project, launched nine years later in 1959, and they’ve been at the forefront of its technological advances ever since.

Machine learning is the discipline in which machines gain understanding.
Machine learning aims to understand, design and use computer programs that can learn from experience (i.e. a data-set) for the purpose of prediction, control or modeling. Think of it this way: we learn what’s right and wrong by going through physical or emotional experiences. It’s kind of similar with computers. The main difference is that their data set uses 0’s and 1’s to learn a lesson, whereas we have to burn our hands to know that fire is hot.

Two great everyday examples of machine learning in action:

  • Netflix’s algorithms learn your preferences by the content that you watch and rate.
  • Image classification or recognition, such as what’s described here by a Facebook engineer.

Machine learning programs are also used for predictive modeling, such as understanding the proliferation of disease, estimating crop yields and anticipating twists and turns in financial markets.

Natural language processing isn’t just for Google searches and bots.
Natural language processing helps machines understand and generate natural language via either speech or text. As consumers, we experience this with automated call centers and Facebook Messenger bots, but there’s so much more that can be achieved with this form of AI.

For instance, companies like Google are using natural language processing algorithms to make ear buds that can translate conversations into other languages in real time.

Robots won’t rule over the world… yet.
For the most part, robots today are designed for a singular purpose or task (remember Boston Dynamic’s Atlas, the back-flipping robot that spurred dozens of memes this past fall?). Those familiar with Parks and Recreation know that consumer robots have not yet reached peak “Rosie the Robot” status from The Jetsons (DJ Roomba, anyone?). However, service robots are beginning to sweep through the hospitality industry by performing simple tasks like bringing up your room service meal once the food is prepared.

The future of robotics is incredibly exciting – and resembles Rosie a little bit, too – in that robots have the ability to both learn and communicate (have you met Pepper?). Imagine a world where you’re driving home in your autonomous Uber pick-up, which tells you that your refrigerator is empty and suggests dinner options from your favorite takeout place.

Ethical problems, right ahead.
The more obvious ethical problem with artificial intelligence is that its proliferation could lead to extensive and pervasive job loss. Check out McKinsey Quarterly’s research and analysis on this subject in, “Where machines could replace humans and where they can’t (yet)” for a sense of the industries and job types that’ll be impacted by AI in the future. They break out the groups into highly susceptible, less susceptible and least susceptible. You can also download the infographic here.

Beyond the job market, there are a variety of ethical problems that arise with these technologies. Consider the potential quandaries that can arise with autonomous vehicles and potential traffic accidents (Who do you save, the driver or the pedestrian?) and algorithmic biases and manipulation as two of the many issues that scientists and engineers are still working tirelessly to better understand and resolve.