
Does Using AI Make Us Stupid?
There is much concern around the introduction of AI into business processes, and one issue is the potential for ‘de-skilling’, or the tendency towards laziness and letting an AI do too much of what humans are supposed to be doing – which is thinking.
Why AI heralds a new age of stupidity – Andrew Orlowski
I have thought a lot about this, and it’s actually not the first time that this issue has come up in my career. I am a draftsman-trained, and I learned technical drafting when drafting machines, tables, and mechanical pencils on vellum were still state-of-the-art. It wasn’t long into my career that CAD (computer-aided drafting) was adopted in a serious way, and there was much concern at all levels of how it was going to influence what we architectural & engineering practitioners produced. There are many similarities between the situation then and what we’re facing now.
In the end, it became clear that the technology was just another tool that allowed good practitioners to produce good work faster and more accurately, and it allowed bad practitioners to continue producing bad work faster. Artificial Intelligence is ultimately going to behave much the same way, I believe, although the impact may reach much farther.
I read the article mentioned above and I do agree with it – conditionally.
If we allow a tool to influence the quality of the primary results of what we produce, that must be done with our oversight. It cannot be allowed to occur on its own without our awareness and input. In the case mentioned in the article, the actual end result of the accountants’ work was being short-cutted, and their personal input, and even their ability to provide that input, was declining. As they determined, that is a short-term gain but a long-term loss, since their people were being ‘disabled’ in the process. Our goal should be to ‘enable’.
Applying this principle to building codes leads me to this awareness: we’re not in the business of producing documentation that is directly derived from building codes – even the drawings we produce are not the actual product. We are, however, in the business of constructing actual buildings that rely on us to understand and interpret building codes (aka ‘rules of engagement’) as they’re applied to the product that we’re actually creating – the building.
If we have a tool or tools that can help us to apply those rules more effectively, more quickly, and more consistently, then I believe that we’re using the tool appropriately and enabling every practitioner involved in the construction of that building to do a better job without taking away the most relevant skills they possess; rather, it allows them to spend more time exercising them.
Let me know your thoughts on this.
Shawn