I’m not one for hot takes. It’s not that I don’t pay attention to ,or have opinions on, current events. It’s just I prefer to keep silent except when I have something to say that hasn’t already been said. So, I wait until others have shared their takes, and when I find that my own opinion differs from those in circulation, and I feel others might benefit from it, I speak (or, more often, I write).
People who don’t know me very well often expect me to have a hot take on whichever current event has everyone’s hair on fire, and invariably I leave them disappointed. When American distance runner Shelby Houlihan was suspended for a doping violation, some random dude reached out to me on Facebook to ask what my take was. I replied that I had no take beyond the obvious, “I hope she’s innocent and I’m disappointed if she’s not.” Never heard from him again. Around the same time, another stranger messaged me to ask for my take on super shoes. I told this person I was neither for them nor against them but confident that democratic processes would sooner or later regulate them in the way that, on balance, was best for the sport. Crickets.
Now along comes ChatGPT, the new natural language artificial intelligence technology that, with minimal direction, can write a reasonably competent essay on just about any subject in a matter of seconds and can also bang out a pretty good endurance training plan in as little time and with as little direction. The other day my friend Jake Tuber shared with me one example of each—an essay and a training plan—in the form of a screen video he’d captured on his computer while monkeying around with the tool. Jake likes to needle me, and I think he was hoping these videos would give me a heart attack by triggering visions of robots stealing my jobs as a writer and an endurance coach. Nice try, Jake!
In all honesty, I confess that my heart did flutter as I watched the videos. But the feeling passed, giving way to deep reflection on the personal implications of this technological leap. Here is one current event that, for me, hits close enough to home to inspire a hot take! So, here goes. . .
As the title of this post indicates, my general take on ChatGPT is that, despite the uncanny potency of the technology, I am not afraid that it will replace me as a writer or as a coach. Judging by the chatter I’ve seen on social media, this perspective is rather common. What’s less common about my perspective, I think, is why I’m not afraid.
It’s been widely noted that there’s nothing particularly creative about ChatGPT’s creations. At its current state of development, the tool can only say things that have already been said before, in different words. It cannot actually say anything novel. But as others have noted, it’s only a matter of time before ChatGPT can do pretty much anything a human writer can do. If you take comfort in saying, “ChatGPT writes well, but I write better,” you’re setting yourself up for ultimate disappointment. Heck, AI has already written a fake Nirvana song that, if it was real, wouldn’t be the worst Nirvana song.
Here’s how I look at it: No matter how sophisticated these tools become, they will never be able to write exactly what I want to write before I get a chance to write it. They may write like me, but they can’t write me. Take my book Life Is a Marathon. In what possible universe could a robot have written that book for me? True, a more sophisticated future version of ChatGPT might be able to take that book and make it slightly better, but it sure as heck won’t be able to replace the experience I had in writing it—the deep concentration, the testing of intellectual limits, the flow states, the breakthroughs and crises of confidence, the inner transformation (as French philosopher Michel Foucault opined, “When I write, I do it above all to change myself and not think the same way as before”)—and this is the other reason I don’t feel threatened by robots that can write.
In short, I write because I love to write and because I have things to say. Nothing that happens in the world around me will ever change that. The world is full of human writers who are superior to me in one way or another. If this alone were a reason for me not to write, I would have quit long ago.
Everything I just said about writing is also true of coaching. No robot can ever coach an athlete exactly as I would coach the same athlete. I will always have something unique to contribute, regardless of how advanced AI coaches become. Plus, I enjoy coaching, and there’s nothing technology can do to change that either. I’m certain that artificial intelligence will change how I coach in the future, but it will never drain the meaning or fun out of my coaching work. Whether readers and athletes still find value in my writing and coaching when robots are really good at both is another matter, but again, there are already plenty of humans who do both things better than I do, so I’m optimistic on this point.
Like Sigmund Freud, I believe that, on balance, technology neither enhances nor diminishes the quality of human life. “If there had been no railway to conquer distances, my child would never have left his native town and I should need no telephone to hear his voice,” Freud writes in Civilization and Its Discontents. I expect that natural language AI tools will prove to be no different. My brother Josh said something prophetic when I asked for his hot take on my hot take on ChatGPT. To paraphrase: Rarely in natural history does one thing replace another. Far more often, the old makes room for the new.
If you’d like to hear a more detailed presentation on this topic, check out tickets for The Endurance Event where Matt spoke about this topic (and hear from 9 other speakers like Ben Rosario and Keira D’Amato about topics their passionate about). Tickets are on sale until February 19th, 2023 and all sessions can be viewed on-demand until February 20th, 2023.