This is a special post for quick takes by aksara. Only they can create top-level comments. Comments here also appear on the Quick Takes page and All Posts page.
As I keep working with LLMs, I feel like it's liberating but also draining.
Liberating in the sense that it's now much easier for me to tinker and 'solve' my curiosity. As a simple example, how to setup so that my Ubuntu machine turns on its WiFi hotspot automatically and figuring out what information flows through my WiFi hotspot[1] . I wanted to explore OpenAI's new signal data[2], and want to skip the not-so-fun part (writing code) to just getting the analysis results. While previously these would have probably taken me a few hours to learn and browse, now it can be done in minutes. It feels liberating that I can lower the barrier and cost of getting new information and knowledge. It satisfies the curiosity part near-instantly.
But at the same time, with more complexities and even more tasks for it: I fear having no capability to review if it did so in a way that is 'proper', I do not prefer to offer my time (capacity) to review it as I prefer to instead ask it to do more things I'm curious about. It feels mistakenly productive, and it feels draining.