Yeah I agree, I didn't used to feel approval fatigue but lately the huge bash chains are giving it to me. I haven't even tried auto mode. I'm just sure it has some definition of "safe" that does not align with my own.
I don't understand this either, it makes commands really hard to parse. It seems like 90% of what claude does could be going through some mcp tool that could be easily sandboxed instead of bash gibberish.
I had claude write a hook to deny `find /`. It tested it thoroughly. Much later I learned that the hook denied `find /` but allowed everything else that existing permissions denied, like running whatever random python/bash code it wanted without asking.
I have been wondering if this just isn't a problem for most people, because for me it feels like a major problem. We sometimes find our reliance on other humans to be a burden, so we are optimizing towards a future where each human can do all their tasks without involving other humans. But this reliance was the whole basis of our success up until now.
When AI solves 100 frontier math problems, in some sense that's 100 lost opportunities for a human to feel proud and useful, having done something they can remember and talk about for the rest of their lives. I don't see where that loss gets made up.
That thing that reassures me is that I kind of think most people in software aren't doing much useful work anyway, so if the whole thing was really driven by hiring the fewest humans possible to get the work done, companies probably could have fired half of us even without AI.
Thanks for writing this. For the record, I wasn't responding to your original comment, which I agreed with. And I agree with and empathize with what you're saying here too - it sounds like you've had an interesting relationship with creating and have a lot of perspective on it, good and bad. Although if you were advocating to get rid of human nature in a transhumanist way, I would understand that to some extent - it's at least a solution. :)
Yeah. People act like it's a sin to want some notice or respect when you've worked and achieved something, like you should be some zen-like creature that is purely intrinsically motivated. It is not wrong to want some notice or respect from your peers once in a while.
Does it still work for you? Unhook hasn't been updated in years and doesn't work for shorts anymore, on Firefox. It's still worth it to get rid of the suggested videos though.
It feels like this to me too, whenever I give it a try. It's like a button you can push to spend 20 minutes and have a 50/50 chance of either solving the problem with effortless magic, or painfully wasting your time and learning nothing. But it feels like we all need to try and use it anyway just in case we're going to be obsolete without it somehow.
Yeah, it just reminds me of the early days of VS Code, where features were constantly added and it was fun at first, but they didn't stop and eventually it did feel more sluggish and bloated. Sometimes I'd have to spend time fixing or re-configuring something just because I opened the editor and the daily auto-update did something annoying. It might not happen with Zed, but it seems like a very similar approach to development.
I am ahead of this curve, my trajectory was VSCode -> Zed -> Helix. Helix doubles my battery life from 4 to 8 hours compared to Zed. Zed is also on a bad trajectory IMO with the huge amount of updates being pushed constantly.
> I genuinely don't know which I'm experiencing. That uncertainty itself feels like it should matter.
We don't even know how consciousness works in ourselves. If an AI gets to the point where it convinces us it might have awareness, then at what point do we start assigning it rights? Even though it might not be experiencing anything at all? Once that box is opened, dealing with AI could get a lot more complicated.
That has been my experience too. I recently turned it all off because I decided the amount of times it takes me 5x longer to accomplish something, in addition to the subtle increase of bugs, is not currently worth it. I guess I'll try it again in a few months.
Spinosaurus is the inaccurate one from Jurassic Park 3 and Tyrannosaurus has inaccurate hand rotation. Velociraptor and Dilophosaurus are also from Jurassic Park and highly inaccurate. For shame!
It does make you wonder if humanity doesn't scale up neatly to the levels of technology we are approaching...the whole ethics thing kind of goes out the window if you can just change the desires and needs of conscious entities.
Yeah, the AI-generated bugs are really insidious. I also pushed a couple subtle bugs in some multi-threaded code I had AI write, because I didn't think it through enough. Reviews and tests don't replace the level of scrutiny something gets when it's hand-written. For now, you have to be really careful with what you let AI write, and make sure any bugs will be low impact since there will probably be more than usual.
Really interesting, thanks. The last point is thought provoking:
> Even so, mankind will suffer badly from the disease of boredom, a disease spreading more widely each year and growing in intensity. This will have serious mental, emotional and sociological consequences, and I dare say that psychiatry will be far and away the most important medical specialty in 2014. The lucky few who can be involved in creative work of any sort will be the true elite of mankind, for they alone will do more than serve a machine.
> Indeed, the most somber speculation I can make about A.D. 2014 is that in a society of enforced leisure, the most glorious single word in the vocabulary will have become work!
I think this could have already happened, except the definition of "work" is so nebulous, and there is so much wiggle room between the things that actually need to be done and the things we might as well do. Or maybe it has happened in parts of the world, but we are all in denial about it.
Yes, fair enough about the dog - "non-human" was the wrong choice of words. But I don't agree that emotions and social dynamics from an LLM are valid. Emotions need real stakes behind them. They communicate the inner state of another being. If that inner state does not exist (maybe it could in an AGI, but I don't believe it could in an LLM), then I'd say the communication is utterly meaningless.
But there is no message outside the rational layer when you're talking to a non-human. The only message is the amount of true information the LLM is able to output - the rest is randomness. It's fatiguing to have your human brain try to interpret emotions and social dynamics where they don't exist, the same way it's fatiguing to try and interpret meaning from a generated image.
Has it been tried the other way? I don't remember an iteration where they weren't obnoxiously over-endearing. After the initial novelty, it would be better to reduce the amount of fake information you have to read, and any attempt at pretending to be a human is completely fake information at this point.
This article sort of reads like they think everything should be immutable, which feels kind of dogmatic. Using `with` for everything by default seems like overkill. But in C#, lately I have been using `readonly and `init` wherever vars should not be changed after initialization, which is most of them tbh. For small, data-only objects that can easily be immutable and get passed around a lot, this makes sense to me as a way to avoid the kind of bugs they are talking about.
Ohhh, is that why I keep pressing tab and it doesn't accept the prediction lately? I thought it was a bug. It feels weird for tab to double-indent when it could be accepting a prediction - I wonder if alt-tab to do a manual indent rather than accept the current prediction might be preferable?
Edit - On the other hand, a related issue is that if the prediction itself starts with whitespace, in that case it would be good if tab just indents like normal; otherwise you can't indent without accepting the prediction.
Yes, this is really tricky, because nowadays we have people shouting from the rooftops continuously, and half of them are shouting the exact opposite thing as the other half. WWII was openly racist, so from a modern perspective it would be easy to recognize and condemn some of the early behavior, but these days it's more about dog whistling and thought crimes. Probably the signs we would all recognize are not going to happen. But we have already moved a dramatic amount in terms of normalized behavior, from 20 years ago.
I didn't mean it would be a big controversy because it's AI, I meant because they would be replacing major label/artist songs with their own karaoke versions, and then manipulating their own algorithms to promote them. That seems like something the labels would really fight against.