Skills for the Future
Barring a major “Mad Max” type scenario (which would anyway make all our hard-earned skills obsolete) recent discussions lead me to believe that one of the key skills in the future will be the ability to communicate concisely. We could call it “cutting through the BS” or “anti-slop vision,” but it appears to me that the future is not likely to belong to those who wield AI like a vorpal sword, puffing up every little idea into the intellectual equivalent of cotton candy, but rather those who use it reluctantly—if at all.1
I am basing this on a fundamental observation, viz., the deluge of AI-based slop I am receiving on a nigh-daily basis, mostly in the form of applications. All of them break the cardinal rule of writing, which states that you must have something to say in order to write. In fact, I like to summarize writing as “simple”2 two-step process:
- Have something to say.
- Say it.
LLMs make it easier for people to clothe their content better, but unless their users show some restraint, LLMs will always err on the side of flattery and purple prose without any additional content. Hence, if you have nothing to say, you may now say it nicer and with more words. Surely, that is precisely what we have been waiting for!
That is not to say that I am against the use of LLMs, on the contrary. But this post is less about how to use LLMs gainfully, a skill that any person with intact critical faculties3 can figure out, and more about a skill that will retain its relevance. Indeed, the ability to read between the lines, perceive the actual content (be it ever so hidden), and strip away everything that is unnecessary may well be the best tool in any person’s intellectual toolbox. Coupled with critical thinking, anyone thus equipped will be able to navigate the world of tomorrow, which, if we are not very careful, will be swamped even more with bogus content.
This is not your standard “cultural pessimism” post; I am actually quite hopeful about the future insofar as slop and low-quality content has always been with us. The fact that we can now mass-produce it is deplorable but also provides ample opportunities to hone one’s intellectual abilities. And who knows, maybe we will soon move beyond inane cotton-candy content and use LLMs in domains to which they can actually contribute?
Here’s to hoping, until next time!
-
Do not let the em-dash fool you. I have been using them long before the advent of LLMs, and, fortune willing, I shall continue using it long after these models have gone the way of the dodo. ↩︎
-
Following the grand tradition of beloved mathematics textbooks here. ↩︎
-
The interesting question is to what extent the critical faculties remain with prolonged reliance on what one may call, somewhat hyperbolically, intellectual crutches. ↩︎