Still thinking about https://lucumr.pocoo.org/2026/1/18/agent-psychosis/
https://www.youtube.com/watch?v=15A0F5aOoPM
Neat!
Stevey's Birthday Blog
31–40 of 40 posts
Re: Stevey's Birthday Blog
#32Good summary. Upshot: Steve thinks he’s built a quality task tracker/work system (beads), and is iterating on architectures, and has gotten convinced an architecture-builder is going to make sense. Meanwhile, work output is going to improve independently. The bet is that leverage on the top side is going to be the key factor. To co-believe this with Steve, you have to believe that workers can self-stabilize (e.g. wit…
> Stay rested Steve, keep on this side of the manic curve please, we need you This is my biggest takeaway. He may or may not be on to something really big, but regardless, it's advancing the conversation and we're all learning from it. He is clearly kicking ass at something . I would definitely prefer to see this be a well paced marathon rather than a series of trips and falls. It needs time to play out.
That something being psychosis
Re: Stevey's Birthday Blog
#33Every time I read another article from this guy I get even more frustrated telling if he’s grifting or legitimately insane Between quotes like these > I had lunch again (Kirkland Cactus) with my buddies Ajit Banerjee and Ryan Snodgrass, the ones who have been chastising teammates for acting on ancient 2-hour-old information. , and trying to argue that this is the future of all productivity while taking time to physic…
> telling if he’s grifting or legitimately insane or if he's talking to us from 5 years in the future. Ignoring the financial aspect, this all makes sense - one LLM is good, 100 is better, 1000 is better still. The whole analogy with the industrial revolution makes sense to me. > AI firms are currently burning money as fast as possible selling a dollar for 10 cents. The financial aspect is interesting, but we're deal…
Better in what sense? What are we actually building with 1000 LLMs
Re: Stevey's Birthday Blog
#34Earlier quoted context omitted.
> telling if he’s grifting or legitimately insane or if he's talking to us from 5 years in the future. Ignoring the financial aspect, this all makes sense - one LLM is good, 100 is better, 1000 is better still. The whole analogy with the industrial revolution makes sense to me. > AI firms are currently burning money as fast as possible selling a dollar for 10 cents. The financial aspect is interesting, but we're deal…
Better in what sense? What are we actually building with 1000 LLMs
A system to build more systems with more LLMs, of course!
Re: Stevey's Birthday Blog
#35Match made in heaven, AI bro turns crypto grifter.
Re: Stevey's Birthday Blog
#36Good summary. Upshot: Steve thinks he’s built a quality task tracker/work system (beads), and is iterating on architectures, and has gotten convinced an architecture-builder is going to make sense. Meanwhile, work output is going to improve independently. The bet is that leverage on the top side is going to be the key factor. To co-believe this with Steve, you have to believe that workers can self-stabilize (e.g. wit…
> Stay rested Steve, keep on this side of the manic curve please, we need you This is my biggest takeaway. He may or may not be on to something really big, but regardless, it's advancing the conversation and we're all learning from it. He is clearly kicking ass at something . I would definitely prefer to see this be a well paced marathon rather than a series of trips and falls. It needs time to play out.
> He is clearly kicking ass at something.
Publishing unmaintainable garbage code to Github?
Have you looked at the beads codebase? It's a bad joke at our expense.
Re: Stevey's Birthday Blog
#37Earlier quoted context omitted.
> Stay rested Steve, keep on this side of the manic curve please, we need you This is my biggest takeaway. He may or may not be on to something really big, but regardless, it's advancing the conversation and we're all learning from it. He is clearly kicking ass at something . I would definitely prefer to see this be a well paced marathon rather than a series of trips and falls. It needs time to play out.
That something being psychosis
And crypto
Re: Stevey's Birthday Blog
#38Every time I read another article from this guy I get even more frustrated telling if he’s grifting or legitimately insane Between quotes like these > I had lunch again (Kirkland Cactus) with my buddies Ajit Banerjee and Ryan Snodgrass, the ones who have been chastising teammates for acting on ancient 2-hour-old information. , and trying to argue that this is the future of all productivity while taking time to physic…
> telling if he’s grifting or legitimately insane or if he's talking to us from 5 years in the future. Ignoring the financial aspect, this all makes sense - one LLM is good, 100 is better, 1000 is better still. The whole analogy with the industrial revolution makes sense to me. > AI firms are currently burning money as fast as possible selling a dollar for 10 cents. The financial aspect is interesting, but we're deal…
> Ignoring the financial aspect, this all makes sense - one LLM is good, 100 is better, 1000 is better still. The whole analogy with the industrial revolution makes sense to me.
How does this make sense? LLMs are doing knowledge work so they face the same coordination problem that humans do, they're not assembly line workers. We have no reason to believe that the lessons of the mythical man month don't apply to LLMs too since the coordination costs, especially when they're touching the same piece of code, are very high.
Re: Stevey's Birthday Blog
#39I think there's an interesting idea behind Gas Town (basically, using supervisor trees to make agents reliable, analogous to how Erlang uses them to make processes reliable), but it's lacking a proper quality ratchet (agents often don't mind changing or deleting tests instead of fixing code) and architectural function (agents tend to reinvent the wheel over and over again, the context window simply isn't big enough t…
Indeed, the Gas-Town token is down 97% from all-time high, see https://coinmarketcap.com/currencies/gas-town/ He's obviously a smart guy, so he definitely should've known better. It's weird how these AI evangelists use AI for everything, but somehow he didn't ask ChatGPT what all of this means and if it may have reputational damage, because I just asked if I should claim these trading fees, and it said: Claiming coul…
> Indeed, the Gas-Town token is down 97% from all-time high,
What else could possibly have happened? Surely every one put their money in with the express intention of participating in a pump and dump.
Not taking the money would have been the high road. I don't think basing the economy on gambling and scams is good for society. But who could realistically claim to be a 'victim' here?
Re: Stevey's Birthday Blog
#40Earlier quoted context omitted.
But why should one expect no future improvement in training costs (all else being equal) from Moore's Law, never mind any future use of eg. more efficient algorithms or more specialised hardware?
If you are quoting Moore's Law in 2026 as the reason LLMs will be profitable, I don’t know how to interact with you. I guess we’ll make up the losses per unit at scale, and grow to infinity.
I'm not making some general claim about LLMs being profitable. To be clear, are you claiming with high confidence that LLM training costs will show no meaningful reduction, for some fixed quality, over roughly the next 10 years?