Heads up, this isn’t one of our typical posts.
While we love sharing news and info about IT services, sometimes we like to just geek out on cool tech.
We are a bunch of engineers, after all.
If you like geeking out on cool tech too, welcome. You are among friends here.
Ok, so what tech are we going to talk about today?
Machine learning refers to computers that have the ability to learn without being explicitly programmed. They grow and change when exposed to new data that they sense from their environments, e.g. a self-driving car that brakes because it detects slow traffic up ahead.
You’ve probably seen on Facebook a video or two of people braving ride-alongs in Google’s self-driving car, or of Alexa, everyone’s personal assistant by Amazon.
While much of the buzz surrounding learned machines has focused on personal applications of the technology, e.g. automobiles and home appliances, we think artificial intelligence’s potential impact on business is the more interesting facet of this topic.
Smart machines have the potential to revolutionize the way companies operate and interact with customers, and in a myriad of ways. It’s impossible to predict all the ways in which intelligent objects will impact business, but here are a few creative ways people have already envisioned:
- Say you’re leasing a (self-driving) car and you missed the last couple payments. The bank can engineer the car to not start until you pay up.
- Robots that do people’s jobs…and we don’t mean, like, ATMs. We mean robots that look like real humans and do real-human things, like answer phones in an office.
- Smart homes that sense the condition of a house (e.g. plumbing, foundation, insulation) and provide this information to potential buyers and mortgage lenders.
We think so.
Alas, despite all the hype, we still have a ways to go before such artificial intelligences are commonplace.
Writers Dorian Pyle and Cristiana San Jose summed it up well in the McKinsey Quarterly, when they wrote, “Dazzling as such feats are, machine learning is nothing like learning in the human sense (yet).”
There are a few, important, kinks to work out.
Security is one. For example, many people have voiced concerns about the susceptibility of self-driving cars. What if a hacker takes over control of your vehicle? Similarly, safety is another concern. What if a self-driving car shuts down on the freeway? And then there’s the privacy issue. For example, many people feel that a car that won’t start if you don’t make timely payments is an invasion of privacy.
These are not insurmountable challenges, but they will take time to address.
Still, as Pyle and San Jose astutely point out, “what [machine learning] already does extraordinarily well—and will get better at—is relentlessly chewing through any amount of data and every combination of variables.”