Are all of our jobs in jeopardy of being taken over by Artificial Intelligence (A.I.)? Well, maybe sooner than we’d all like to think.
Laura Sparks, of Creative Sparks Writing, presented, “Don’t Panic – The Role of Human Writers and Editors in the World of A.I.,” hosted by the Ohio Editorial Freelancers Association on January 26. Her presentation was insightful, thought-provoking, and more than a little worrisome, to say the least.
According to Sparks, there are different types of business applications that currently use machine-generated content, or A.I. Some recognizable examples, are:
- Using metrics to resell unused tickets on sports sites.
- Writing product descriptions on landing pages.
- Producing financial reports.
- Producing Call Center reports.
- Writing property descriptions for real estate agencies.
- Producing Fantasy Football leagues.
So how does A.I. work? First, it needs a data source. Second, a template. Third, it uses complex formulas and natural language processing to identify trends within the data to generate it into narrative.
A.I. can analyze data very quickly, much faster than a human can. It can generate an entire whitepaper literally within seconds. A.I. can even imitate humor. Additionally, Gmail is now getting ready to roll out sentiment analysis. Sentiment analysis can suggest possible responses to an email chain. When texting on your phone, it will be able to give you possible responses to answer back. It is even smart enough to know, due to complex analysis, that when two people are arguing via email, responding with “I love you,” will diffuse the argument.
Although impressive, there are just some things that A.I. cannot do.
It cannot accomplish face-to-face interviews or even telephone interviews. It can’t select the right angle to use from an interview, nor can it determine variations in language from different parts of the country. It can’t decipher dialects or recreate human intuition, or know what is socially acceptable and what is not. It has no sensitivity, knowledge of diversity or ethics. It can be hacked and not even realize it.
It can’t yet be programmed to recognize empathy or sympathy, but data analytics are now being explored to use as a foundation for that understanding, and one day, may use complex algorithms to accomplish this.
This evolution has begun to take on speed, though, and although there is a lot that A.I. is not capable of at this point, it doesn’t mean that it won’t be in the near future.
Sparks refers to the great strides Google Translate has made with the use of neural networking. She suggests reading “The Great A.I. Awakening,” by Gideon Lewis-Kraus, published by the New York Times on Dec. 14, 2016, to get a better view of the “evolution” that is happening with regard to machine learning. This new A.I. phenomenon described as “a new computational platform” is backed by the commercial institution known as Google Brain.
One of the hopes of Google Brain is that computers, at some point, would evolve to the point where “advanced skills would emerge organically,” and posits that the computer “wouldn’t need to be pre-programmed with fixed rules. It would, instead, rewire itself to reflect patterns in the data it absorbed.”
According to Lewis-Kraus’ article, in earlier versions of A.I., when the computer was able to beat the reigning Chess champion, that was impressive, but then when the computer could beat the current Go champion, that was much more impressive. Getting good at solving problems with what adults would do based on firm rules based in math was one thing, but to be able to teach a computer to learn to solve the problems like a child, when no foundational rules were in place, was, and is, much more complex, and is a huge leap forward.
It’s probably just a matter of time before this evolution sees fruition, as the top technological companies in Silicon Valley race to be the first to make that happen. “Google, Facebook, Apple, Amazon, Microsoft and the Chinese firm Baidu” all crawl over each other to find the top talent to make it so, even reportedly paying “seven-figure salaries to the most promising graduate students,” according to Lewis-Kraus.
And we all have heard of software applications such as Acrolinx, which can accomplish machine editing quickly and pretty accurately — again, not to the point that human intervention is not needed, but good enough that it saves a lot of editing time and can allow writers to virtually become their own editors.
So, is the “A.I.” Boogie Man, then, a real threat to writers and editors, today? Maybe not today, but as we all know, tomorrow always comes, whether we want it to or not.
For additional information about the evolution of A.I., machine-generated content and machine learning, Sparks recommends these articles:
- Robot journalist writes article on LA earthquake:
- Don’t panic, but this article was written by algorithm
- Are robot writers really about to take over the world?
For more information about the Editorial Freelancer’s Association, go to http://www.the-efa.org/.
By Lynn Nickels