> Here's a thought. Lets all arbitrarily agree AGI is here.
A slightly different angle on this - perhaps AGI doesn't matter (or perhaps not in the ways that we think).
LLMs have changed a lot in software in the last 1-2 years (indeed, the last 1-2 months); I don't think it's a wild extrapolation to see that'll come to many domains very soon.
Although there is the consistent trap of tools that assign threads/workers based on the number of cores (e.g unit testing or bundling tools). This means the efficiency cores get dragged in and can absolutely tank the process.
This was particularly pronounced on the M1 due to the 50/50 split. We reduced the number of workers on our test suite based on the CPU type and it sped up considerably.
It used to be Factorio for me (I live in Australia, so long flights happen a lot). The problem with Factorio the flight isn't long enough! and the game bleeds into 100+ hours post-flight.
DF gets all the news (rightfully so, it's an epic game that I've dumped a ton of hours into) but if you haven't already, consider checking out Songs of Syx. It's like DF but multiplied by 100. You can have tens of thousands of citizens, doing most of the things they do in Dwarf Fortress, and a lot more, including waging huge wars against the neighbors. The limits of DF kinda made me sad, actually, that you are limited to so few Dwarves (and don't say it's because you want to know the story of all of them, because after 30 or so you lose track of who is who anyways, so might as well up the limit from 100 to 50K, or more? ;) Songs of Syx has also routinely been getting massive updates since 2020 and I have a feeling the code is a bit cleaner so the solo dev can add features faster (unlike DF's code base which is, according to one of the new devs a nightmare to work with). It's a game that is never talked about but deserves a whole lot more love from gamers.
I don't mean to cast shade on DF, I really do love it, and am happy for its existence, I just think that DF fans should also look into Songs of Syx.
The defining difference for me are the generated stories in DF, which often are a lot of random trash but still give a feeling of a deeper meaning.
I’ve definitely experienced a subjective regression with Opus 4.5 the last few days. Feels like I was back to the frustrations from a year ago. Keen to see if 4.6 has reversed this.
> It's so interesting to watch an agent relentlessly work at something. They never get tired, they never get demoralized, they just keep going and trying things where a person would have given up long ago to fight another day. It's a "feel the AGI" moment to watch it struggle with something for a long time just to come out victorious 30 minutes later.
This is true... Equally I've seen it dive into a rabbit hole, make some changes that probably aren't the right direction... and then keep digging.
This is way more likely with Sonnet, Opus seems to be better at avoiding it. Sonnet would happily modify every file in the codebase trying to get a type error to go away. If I prompt "wait, are you off track?" it can usually course correct. Again, Opus seems way better at that part too.
Admittedly this has improved a lot lately overall.
I don't understand why anyone finds it interesting that a machine, or chatbot, never tires or gets demoralized. You have to anthromorphize the LLM before you can even think of those possibilities. A tractor never tires or gets demoralized either, because it can't. Chatbots don't "dive into a rabbit hole ... and then keep digging" because they have superhuman tenacity, they do it because that's what software does. If I ask my laptop to compute the millionth Fibonacci number it doesn't sigh and complain, and I don't think it shows any special qualities unless I compare it to a person given the same job.
You're a machine. You're literally a wet, analog device converting some forms of energy into other forms just like any other machine as you work, rest, type out HN comments, etc. There is nothing special about the carbon atoms in your body -- there's no metadata attached to them marking them out as belonging to a Living Person. Other living-person-machines treat "you" differently than other clusters of atoms only because evolution has taught us that doing so is a mutually beneficial social convention.
So, since you're just a machine, any text you generate should be uninteresting to me -- correct?
Alternatively, could it be that a sufficiently complex and intricate machine can be interesting to observe in its own right?
If humans are machines, they are still a subset of machines and they (among other animals) are the only ones who can be demotivated and so it is still a mistake to assume an entirely different kind of machine would have those properties.
>Other living-person-machines treat "you" differently than other clusters of atoms only because evolution has taught us that doing so is a mutually beneficial social convention
Evolution doesn't "teach" anything. It's just an emergent property of the fact that life reproduces (and sometimes doesn't). If you're going to have this radically reductionist view of humanity, you can't also treat evolution as having any kind of agency.
"If humans are machines, they are still a subset of machines and they (among other animals) are the only ones who can be demotivated and so it is still a mistake to assume an entirely different kind of machine would have those properties."
Wrong level of abstraction. And not the definition of machine.
I might feel awe or amazement at what human-made machines can do -- the reason I got into programming. But I don't attribute human qualities to computers or software, a category error. No computer ever looked at me as interesting or tenacious.
Yes + Appears it's a rigid structure w/ the engine pushing from the back? At 0.1g I suspect even with advanced composites only a few km would be possible.
This is Windows Aero all over again - why is this a persistent design?
You can't see or process the information behind the glass - at best it's major cognitive load to do so, at worst it's just very noisy with zero added information.
A slightly different angle on this - perhaps AGI doesn't matter (or perhaps not in the ways that we think).
LLMs have changed a lot in software in the last 1-2 years (indeed, the last 1-2 months); I don't think it's a wild extrapolation to see that'll come to many domains very soon.
reply