In 2019, I had my first encounter with an AI model that could write from a prompt. GPT-2 had its flaws, but still was remarkably coherent. It was terrifying.
Fast forward a few chaotic years to 2022, when my attention was elsewhere, most of it online behind a Zoom screen or isolated, as so many of us were. For a class project I revisited the GPT-2, only to find that it had morphed into a far more sophisticated set of models. By the end of the year, GPT-3 had taken the headlines by storm.
Change is inevitable, but that doesn't make it any more likable, and certainly no easier to navigate. This blog is my space to try to sort out this emerging phase in digital life. As an educator, I have to figure out what it means to coexist with GPT-3 and its analogs, since they aren't going away anytime soon. As a writer, I also have to contend with one more form of "competition" that cheapens the value of my skill set. As someone who thrives on flux and creative possibilities, I am also thrilled and fascinated by this latest disruption.
The takeaway: AI language models aren't going anywhere. Let's learn more about them.
Commentaires