Should you fear AI?

With the recent public launch of ChatGPT (and Dall-E earlier), Artificial Intelligence (AI) is going mainstream, with some of its power now available for the average person to use. I must confess that my initial whirl with ChatGPT was purely for fun. Asking ChatGPT to complement my meager creative skills to quickly generate poems, sonnets, limericks and the like, to first amaze, then amuse my friends and family. The results ranged from modestly creative to the very pedestrian, but given the barebones prompts I provided and how quickly the output came back, impressive nevertheless.

What these public releases have also prompted is the enduring debate about the dangers of AI. For those of us old enough to remember HAL 9000 or SkyNet, there is a genuine concern about how much AI will evolve and when it might attain consciousness and hence, rival our own survival instincts. Better minds than mine are debating this, so we will focus on the near-term implications of AI as we know it, and how it might affect our work. And specifically, the work we do at Nayamode/Bluewave — a creative agency that creates various types of content for Fortune 50 brands.

From what we see thus far, the creative abilities per se of ChatGPT are fairly rudimentary. If anything, you could argue that it’s not truly creative in that it generates text/copy based on vast amounts of content that has been fed into it. i.e. it has access to millions of pieces of existing content and has ‘learned’ from them on what appropriate structures are for the different types. And it has algorithms that help determine what is the most likely word to follow from a previous one — again, based on its gargantuan repository of existing material created by humans.

What it is clearly capable of doing very well, and way faster than any human today, is to access this seemingly endless trove of material and to pick relevant pieces. And in what appears to be a huge leap forward, it has the ability to understand very conversational prompts and return content that is equally ‘human-like’, making the user experience more natural and friendly.

These are significant abilities when you consider that oftentimes the work that goes into creating technical content is in large part about researching, gathering, and ordering relevant material, followed by the ability to synthesize it in a manner that makes the final product more than just an amalgamation of information or data. This last part very much requires a human’s effort but the initial tasks (which can often be a majority of the effort) can likely be done well by ChatGPT.

What are the implications of this? Subject Matter Experts (SMEs) and high-level content creators don’t need to fear ChatGPT. Not yet.

At this point, we see ChatGPT (or similar) augmenting the efforts of a human content creator in a manner that allows them to scale, be more efficient and productive. This does have downstream effects though — specifically, that while content volume is likely to go up, the number of human creators needed for creating it will go down.

Net — fewer creators will create more high-quality content with the aid of ChatGPT and its like. Which will almost necessarily mean that some lower-level creators will be rendered redundant. Perhaps not terribly different from what technology has done to scores of professions over the last few decades.

Ten years from now though, as ChatGPT gets increasingly proficient and ‘intelligent’, we wouldn’t bet on content writers’ professions being secure. If you’re already in the field, best to enhance your skills today, keep figuring out how to stay a step ahead of the machine and ride out the wave for as long as you can. If you’re just entering it though, may we recommend you look elsewhere that is a bit more future proof?

Leave a Reply

Your email address will not be published. Required fields are marked *

15 + 8 =