It is natural a to want to do more with less. Over the past few centuries we have relied more on machines to do a lot of our work for us. It was once thought that computers would help us do all of our work and we would sit by at our leisure and enjoy life. We are enterprising creatures, looking to get the most value out of the least amount of work.
It’s human nature.
As you know, that utopian world of non-working people and hard working machines hasn’t come to pass, but it won’t stop us from trying.
We Are Relying More on Systems
A couple of seemingly disconnected things have come together over the past few weeks to make me reimagine this world of machines anew.
First, I read Maddie Grant and Jamie Notter’s book Humanize, which made me consider how our companies have taken on the likeness of machines, and as such have lost some of their flexibility and adaptability.
Second, I heard about a new computer program that can write articles that are just as convincing, if a little less inspired, than human journalists. Read this article “Can an Algorithm Write a Better News Story Than a Human Reporter?”
Third, I attended the Social Media Analytics Summit, where one of the arguments were around the idea that automated sentiment analysis is flawed and businesses should not use it to make decisions. Here is a really great post that explains why automated sentiment analysis is not very accurate.
Fourth, I heard a segment on NPR that pointed out a new computer program that can grade student papers more accurately than the teachers. Good news for over busy teachers, but maybe not so good for creative thought. Check out the story, “Can A Computer Grade Essays As Well As A Human? Maybe Even Better, Study Says.”
Finally, I heard about an experiment using a robot nanny. From the article, “80 percent [of kids] thought the robot was intelligent, while 60 percent thought it had feelings….The children didn't believe that the robot had the right to be paid to work or vote in an election, and most felt that it was okay for it to be bought and sold.”
Taken separately, these things are just a great book, a concern about a type of analysis and interesting news stories. But in its totality, it got me to thinking about how we turn over more of our decision-making power to machines every year and how easy it has been for us (me included) to relinquish that power.
In our rush to increase productivity and get simple answers, we have reduced people down to single digit Klout scores, judged people by their avatar, and valued companies by their digital footprints.
The Digital Migration
The recent squabbles over the valuation of Facebook as it nears its IPO on one hand and the sale of Instagram for $1 billion US to Facebook show that our society has certainly moved online for good. Perhaps our collective move away from manufacturing and toward a service and knowledge-based economy have made this mass migration into the digital world inevitable.
After all, as humans we have been migrating since the start of time. Why should this be any different?
It seems that the sheer speed of this transition, more than any other in human history, is moving faster than we are adapting our laws, our social understanding or even our brains.
In an effort to “keep up” we turned over the mundane tasks to computers. This is all well in good, but now a “little” has turned into “a lot.” We regularly let machines determine what times are best to tweet (Buffer), what words to use (Scribe SEO or Inbound Writer), or even what products to buy based on our values (Good Guide). Some of use use the Internet a little bit like we would an external brain (witness my Diigo account).
We are seeing the world through our own self-imposed , filtered glasses, but we forgot to take out the rose colored lenses.
The Trouble with Machines
There is so much good that comes with machines that it is hard to see the drawbacks, but here are six areas I think are impacted by machines. What are the drawbacks that you see?
- Lack of Thoughtfulness. Sure a computer can consider many options, as long as they has been programmed into the system. A human is much better at processing information and seeing the subtlety. Moreover, we think more critically than a machine can do.
- Lack of adaptability. While it is true that neural networks can learn and “adapt” to new input, they can only consider the variables input into the system. What about information that is missing, body language and other clues for which it is very hard for a machine to understand or react.
- Loss of Creativity. If we rely on machines to relay information, there is not the same room for the creative turn of phrase. And what about breaking the rules of grammar to make a point. Get my point? The use of machines to set the standard for writing sets the bar too low.
- Autopilot Mistakes. When a system is on autopilot, even small deviations can eventually lead to huge accidents and unintended consequences. Machines can make recommendations and do the heavy lifting in a quicker time and with more accuracy than a human, but it isn’t as good at making decisions that require opinion.
- Spam. With the ability for machines to write for humans more convincingly, it leaves the door wide open for more spam sites and email. If humans don’t even have to write the stuff it becomes even more prevalent.
- Shallow Thinking. Some scientists and thinkers have hypothesized that our brains are being rewired to only have shallow thoughts. Even if you don’t completely buy that hypothesis (I don’t), it is true we have become a generation of skimmers. The sheer volume of information thrown at us every day necessitates it. More volume can only exacerbate the problem.
No comments:
Post a Comment