The Art of Kicking

I recently listened to an episode of the Dwarkesh podcast titled "Sholto Douglas & Trenton Bricken - How to Build & Understand GPT-7's Mind". The host and namesake of the podcast, Dwarkesh Patel interviews two of his machine learning friends. While the main topic is the future of AI, the two guests also talk about their personal lives and how they managed to land jobs at big companies and make an impact despite not being in the field for very long.

Around halfway through the podcast, Dwarkesh mentions that he think people are often falling prey the Sunk Cost Fallacy and stay on a path even though they know there is a better option for them. Based on this, the following conversation ensues.

Trenton Bricken 01:38:04
I think it's, “strong ideas loosely held” and being able to just pinball in different directions. The headstrongness I think relates a little bit to the fast feedback loops or agency in so much as I just don't get blocked very often. If I'm trying to write some code and something isn't working, even if it's in another part of the code base, I'll often just go in and fix that thing or at least hack it together to be able to get results. And I've seen other people where they're just like, “help I can't,” and it's,”no, that's not a good enough excuse. Go all the way down.”

Dwarkesh Patel 01:38:36
I've definitely heard people in management type positions talk about the lack of such people, where they will check in on somebody a month after they gave them a test, or a week after they gave them a test, and then ask, “how is it going?” And they say, “well, we need to do this thing, which requires lawyers because it requires talking about this regulation.” And then it’s like, “how's that going?” And they’re like, “we need lawyers.” And I'm like, “why didn't you get lawyers?”

Sholto Douglas 01:39:02
I think that's arguably the most important quality in almost anything. It's just pursuing it to the end of the earth. Whatever you need to do to make it happen, you'll make it happen.

While listening to this, something clicked in me. I realized that the person who is just idling because no one told him what exactly to do next is me. Often, when I am working on a project and am hitting a roadblock that I can't solve with a Slack message, I just abandon it. In fact, I had a situation just as described a few months ago, where it was unclear what regulations I had to abide by. I asked my supervisor who didn't have an answer either, and, instead of contacting a law firm and getting an estimate on what an answer would cost, I just did nothing. Since then, several other research groups, who either figured out the regulatory issues or ignored them, published papers on the topic. I could have done the same, except that I'm not good at kicking myself.

This is something, I want to change. I think that general awareness of the problem will already go a long way, but might not be enough to overcome the problems where I need to kick myself very hard. For that, I am planning to include something like "Kicking-myself-Saturdays" into my next productivity plan. This would mean that I collect all the tasks that I am avoiding, but are not super time-intensive and work through all of them before I am allowed to use my phone for internet browsing.