3 ms·
"running yourself over with a car" feels very different from "learning to use LLMs to your advantage".
by rrr_oh_man 2y ago
"running yourself over with a car" feels very different from "learning to use LLMs to your advantage".
- _heimdall 2y agoThe GP was pointing out that learning to use an LLM, in their opinion, would stop them from learning other new skills and erode their existing ones. In that context I think the analogy holds. Using an LLM halts your learning, as does running yourself over with a car. It's an exaggerated point for sure, but I think it points to the fact that you don't have to learn to use LLMs simply because it's a skill you could learn, especially if you think it will harm you long term.
- fragmede 2y agowhat's better, I stare at my code for 3 hours to find a missing semicolon, or ChatGPT explains it to me in minutes. what lesson am I learning by staring at my code?
- _heimdall 2y agoThe skill of finding errors is actually a very useful one. I can count on one hand how many times I've made similar mistakes in code, after 3 hours staring at it that lesson really sticks and you don't make it again. When an algorithm tries to fix it for you you only learn how to send the code to the algorithm to fix it for you. It's convenient in the moment for sure, but you haven't learned how to fix your own code and won't know how to fix it when the algorithm is the cause of the error. That also ignores secondary risks, like giving your codebase to a nonprofit / very much for profit company. That isn't always possible depending on the codebase, and in general why bother giving them access to it when you instead learn how to find your own missing semicolons? Why spin up a pipe of GPUs and burn all that power rather than learning to do it yourself?