We saw some troubling artificial intelligence (AI) coverage recently. There was an artist who won an art contest with AI-created art, and folks were screaming foul. Then we had concerns about how students could use AI authors to write papers. They wouldn’t be caught by using plagiarism tools, because they were original work, but written by the AI.
This is a recurring problem with any new technological tool. When it comes out, people complain that the tool, particularly for students, should be banned, even though, once the student leaves the education process, the skill of knowing how to use that tool will be more valuable than the skill of producing something without it.
Let me explain.
Training for the future vs. the past
Education should be focused on training students for the jobs that will exist when they graduate, not the jobs that existed before they entered school. Yet, education changes slowly.
When I went to work for a large tech company, schools didn’t yet have PCs, the company wasn’t yet using modern word processing, and the company’s productivity was crippled, because it seemed unable to quickly pivot to the newer tools. The PCs we had were also deployed with terminal emulators, which made them slow to boot and painful to use.
Looking back, what the company should have done was pivot hard to the PC model and move into the current decade, but it didn’t. That is far from an uncommon reaction to new technology. We tend to resist change, and as a result, we often do not benefit from the technologies that could improve our competitive posture.
The products highlighted above, both AI art creation and paper writing, can, if they are used properly, turn poor writers and those without art skills into decent writers and artists.
They can allow us to use existing human resources to do new and interesting jobs for which we were previously unqualified. The effort to develop a concept and refine it through multiple drafts as opposed to trying to do it right the first time leads to what should be higher overall quality across a larger variety of employees and potentially makes people more productive.
The employees of the future may be more valuable if they have strong AI interface skills, much like strong internet search skills are valuable today for certain jobs, but learning Boolean logic was not a part of most school systems.
Just like internet search, to be efficient requires some understanding of Boolean logic. So, too, does working with an AI tool to build an image or a document require an understanding of how to efficiently tell an AI what you want. The less the user knows about this process, the longer it will take to achieve an acceptable result, which reduces productivity and lowers job satisfaction by introducing frustration.
The problem is there do not appear to be any courses to train users on how to work with an AI. There are plenty of courses on how to build an AI, how to create a conversational AI, and how to train an AI but not how to use one. Granted this is because personal AIs, while proposed by companies like IBM, are not yet at critical mass. But much like we are building skills for quantum computing long before we have quantum computers, we should be encouraging the development of skills needed to work with personal AIs before people get them.
Train for benefits
Education lags technology by decades, which creates a significant problem with new and existing employees’ ability to make early and best use of these expensive but potentially revolutionary new tools.
Consider, as you bring your own AI tools into production that you also need to assure the users are trained in their proper use.
Or the full benefits of these tools will not be realized until that training becomes predominant.