When one thinks about human decision making, there are at least two classes of decisions:
1. decisions made with our "fast" minds: ducking out of the way of an incoming object, turning around when someone calls our name ... a whole host of decisions made without much if any conscious attention, and that if you asked the human who made those decisions you wouldn't get much useful information about.
2. decisions made with our "slow" minds: deciding which of 3 gifts to get for Aunt Mary, choosing to give a hug to our cousin, deciding to double the chile in the recipe we're cooking ... a whole host of decisions that require conscious reasoning, and if you asked the human who made those decisions you would get a reasonably coherent, explanatory logic chain.
When considering why an LLM "made the decisions that it did", it seems important to understand whether those decisions are closer to type 1 or type 2. If the LLM arrived at them the way we arrive at a type 1 decision, it is not clear that an explanation of why is of much value. If an LLM arrived at them the way we arrive at a type 2 decision, the explanation might be fairly interesting and valuable.
You're really rizzing up the whole "AI can do almost everything better than humans" point. Is there a chance that your investments are causing you to sensationalize things a bit? Because I can promise you, AI can only do better the things that I have absolutely no skill in.
It's an interesting one. We'll have to discover where to draw that line in education and training.
It is an incredible accelerant in top-down 'theory driven' learning, which is objectively good, I think we can all agree. Like, it's a better world having that than not having it. But at the same time there's a tension between that and the sort of bottom-up practice-driven learning that's pretty inarguably required for mastery.
Perhaps the answer is as mundane as one must simply do both, and failing to do both will just result in... failure to learn properly. Kind of as it is today except today there's often no truly accessible / convenient top-down option at all therefore it's not a question anyone thinks about.
We want machines to do the laundry and clean our house so we have more time to create art and write code. Seems like in our current trajectory, the machines will produce the art and code so we have more time to clean our house and do laundry....
This is a great framework for self-development, but I wonder if the Job vs. Gym analogy is a bit premature. There seems to be a level of Silicon Valley optimism here that assumes AI already surpasses human capability in these creative areas. From my perspective, AI only outperforms in areas where the human hasn't developed a real craft. Is it possible that the current hype is causing us to undervalue the unique quality of human-only output?
- using ai effectively is itself a skill that needs training, especially if you're already good at the critical thinking stuff.
- using AI actually most of everything i do anyways, is critical thinking. I'm constantly reviewing the AI's code and finding little places where the AI tried to get away with a shortcut, or started to overarchitect a solution, or both.
The point that we should do the things we want to be, as in, to have as parts of our identity, is really good insight. Even if the AI can do X well, maybe I want to also be able to do X, therefore I should practice it.
I don't know what those things will be for me, yet, but it's good to have a more specific and directed way to think about which skills I want to keep.
Lurking between the lines in arguments about AI writing/code/art is that whether or not an activity is "gym" or "job" is often in the eye of the beholder.
People who never "went to the gym" in a field are all too eager to brush off the entire design space as pure Job that can and should be fully delegated to AI posthaste.
This essay falls under a growing category I call "start off sounding like you agree with critics of GenAI, then subtly attempt to undercut all of the power of their arguments."
The best way to engage with these sorts of articles is to completely ignore all stated advice and move on. There is no "separation of moving things on the job vs. moving things at the gym" when it comes to creative craft, the entire analogy is completely absurd.
10 comments of 43
[ 3.9 ms ] story [ 35.3 ms ] threadWhen one thinks about human decision making, there are at least two classes of decisions:
1. decisions made with our "fast" minds: ducking out of the way of an incoming object, turning around when someone calls our name ... a whole host of decisions made without much if any conscious attention, and that if you asked the human who made those decisions you wouldn't get much useful information about.
2. decisions made with our "slow" minds: deciding which of 3 gifts to get for Aunt Mary, choosing to give a hug to our cousin, deciding to double the chile in the recipe we're cooking ... a whole host of decisions that require conscious reasoning, and if you asked the human who made those decisions you would get a reasonably coherent, explanatory logic chain.
When considering why an LLM "made the decisions that it did", it seems important to understand whether those decisions are closer to type 1 or type 2. If the LLM arrived at them the way we arrive at a type 1 decision, it is not clear that an explanation of why is of much value. If an LLM arrived at them the way we arrive at a type 2 decision, the explanation might be fairly interesting and valuable.
It is an incredible accelerant in top-down 'theory driven' learning, which is objectively good, I think we can all agree. Like, it's a better world having that than not having it. But at the same time there's a tension between that and the sort of bottom-up practice-driven learning that's pretty inarguably required for mastery.
Perhaps the answer is as mundane as one must simply do both, and failing to do both will just result in... failure to learn properly. Kind of as it is today except today there's often no truly accessible / convenient top-down option at all therefore it's not a question anyone thinks about.
- using AI actually most of everything i do anyways, is critical thinking. I'm constantly reviewing the AI's code and finding little places where the AI tried to get away with a shortcut, or started to overarchitect a solution, or both.
I don't know what those things will be for me, yet, but it's good to have a more specific and directed way to think about which skills I want to keep.
People who never "went to the gym" in a field are all too eager to brush off the entire design space as pure Job that can and should be fully delegated to AI posthaste.
The best way to engage with these sorts of articles is to completely ignore all stated advice and move on. There is no "separation of moving things on the job vs. moving things at the gym" when it comes to creative craft, the entire analogy is completely absurd.