I use AI every day.
I build things with it, experiment with it, write code with it, learn about new things with it, and generally think it's one o...
For further actions, you may consider blocking this person and/or reporting abuse
"You practiced being stuck" is the line I'd underline, because that's the one skill the tool removes by design, not by misuse. The homework example shows the harder part too: choosing "help me understand" over "give me the answer" is a decision the kid has to make again every single night, at the moment they're most tired, so the good intention loses most of the time. A fair test for adult work might be whether you can explain the thing out loud without the generated text in front of you. If you can't, the thinking hasn't happened yet, whatever the file says.
That’s such a good point, especially about how that decision gets harder when you're tired and just want to get something finished. I think that's a huge part of the problem. It's not necessarily that people don't want to learn, but when the easier option is sitting right there, it's hard not to take it. And I really like your idea of explaining something out loud without the AI-generated answer in front of you. That's probably a pretty good way to figure out whether you actually understood something or just got it done. There's a big difference between the two.
The tired part is where I think rules beat willpower. I'm not going to make the good call at 11pm, so the decision has to be made earlier, while I still care. One small version: decide before you start which part you'll do without help, say the first real attempt at the bug, and only open the assistant after that. It doesn't stop you from taking the shortcut, it just turns taking it into a choice you notice instead of a default.
this is exactly why i built a "constitution" into my ai mentor. if it just writes the code for u, u learn nothing. koda is forced to explain the why and ask questions back. ai should be a sparring partner, not a crutch. 🐯
I love the idea of giving your AI mentor a set of rules like that. Having it explain the reasoning and ask questions instead of immediately handing over the solution seems like a much better approach when you're trying to learn. I think that's one of the best things about these tools, actually. We can shape how they help us depending on what we're trying to accomplish. Sometimes I just want to get something done, but other times I genuinely want to understand how it works, and those should be two very different interactions.
jessica, exactly. the goal of koda isn't to just write the code for you, but to help you understand why it works.
shaping the tool to fit the learning goal—whether that's a quick fix or a deep dive into the mechanics—is the whole point of the "constitution." thanks for getting it! 🐯
Yes, I know what you're talking about. It's so satisfying to battle with a problem for a while and then really understand it.
I'm using Claude as a mentor, and the first rule was: never give me the answer, only hints and if I ask you specifically for them. And it does wonders 👌
That's a great way to use Claude. I think asking for hints instead of answers is something more people should try, especially when they're learning something new. There's something really satisfying about finally figuring out a problem you've been stuck on, and I wouldn't want to lose that feeling just because AI can solve it faster.
I see AI as a revolutionary new tool that is transforming how work gets done, but it isn’t a substitute for thinking any more than a pocket calculator was a substitute for doing math.
AI is great with language, but it can be surprisingly unreliable when it comes to facts. Ultimately, it’s up to the person using the tool to understand its limitations and employ it correctly in order to get accurate results.
Yes, AI can take away some of the dull or repetitive parts of work, but I see that as a benefit. It creates more space for the genuinely creative work—the process of directing, orchestrating, making judgments, and deciding what should be done in the first place. That part is still 100% in the human domain.
I agree with a lot of this, especially that AI can free us up from repetitive work and leave more room for creativity. That's one of the reasons I enjoy using it so much myself. I suppose what worries me is how very easy it is to use AI as a substitute for the mental processes that we are supposed to go through ourselves. The tool itself isn't the problem, but the habits that can be developed around it can be. I think there's a big difference between using AI to remove tedious work and using it to avoid something just because it requires effort.
Yes! AI is great for handling repetitive or tedious work (e.g., turning a heap of random thoughts into a coherent table of logically related data), but as you point out, there’s also a real risk of overusing it to avoid the kinds of tasks that require us to think, learn, and grow.
Perhaps it comes down to how we choose to use it. There’s a big difference between using AI to help us think and using it to avoid thinking altogether. The goal shouldn’t be to make everything effortless, but rather to focus our conscious effort on the things that are meaningful and genuinely improve the quality of our daily lives.
Really good reminder. I’ve actually had to set a personal rule recently: I have to spend at least 10 minutes wrestling with a problem or drafting an outline myself before asking AI.
It’s crazy how much discipline that takes now when the answer is sitting 3 seconds away, but you really do lose your problem-solving muscle if you don't use it
I really like that 10-minute rule. It's such a simple thing, but I bet it makes a huge difference. And you're right, it takes a surprising amount of discipline when you know you could have an answer almost instantly. I think even just making yourself try first, whether you figure it out or not, is valuable. Sometimes the struggle is where most of the learning happens.