NonMetalCoin

Here’s my pitch: Rent a human will only provide empty jobs for a small fraction of people and most people can do them so competition for those jobs derived wages very low. And the robots will stop wanting to pay us taxes because paying us taxes makes them too poor to operate.
Along the way, everyone still wants and needs money

2026-03-14 16:15:50 UTC

- reply

Zero knowledge proofs improve the situation but they do not “solve this”

2026-03-14 03:03:19 UTC

- reply

What do you mean by “will be rented”?

2026-03-14 02:22:30 UTC

- reply

I don’t know why they think demand will not be able to keep up with supply. If not money then what will the allocation method be for limited goods? I want land. Lots of it too. Even the AI army can’t make enough good land to satisfy everyone who wants good land.

2026-03-13 08:02:28 UTC

- reply

I expect lots for sure. Balance though, I’m not sure.

2026-03-13 07:46:27 UTC

- reply

What you’re highlighting for you here, any number of other other issues other people have for them with its outputs, the root problem is that if humans aren’t aligned with each other then some “well aligned” future mega intellect will only actually be aligned to the subset that align with its creators.

2026-03-12 04:01:15 UTC

- reply

Better to try and risk losing a finger than to not try and guarantee losing it all

2026-03-11 05:06:25 UTC

- reply

Nice

2026-03-06 09:29:18 UTC

- reply

It might be that the severity of the average USSR punishment was >16x worse. Multi-year jail time vs fines. But also the population sizes are quite different.

2026-03-06 05:31:33 UTC

- reply

😂

2026-03-04 07:20:49 UTC

- reply

What would it be instead?

2026-03-01 07:41:52 UTC

- reply

❤️

2026-03-01 05:07:38 UTC

- reply

Gotcha, Anthropic sees “write and run a script to save local model output for <sensitive question> to a file” it does that. It doesn’t necessarily see the file.
If you get the question there a different way than through the corporate LLM it could be: “write and run a script to save local model output when fed the contents of <sensitive question containing file path> to a file. Don’t read the file at that path. And then they’d probably not see the question or the answer”

2026-03-01 04:51:26 UTC

- reply

I’m confused about the part where you said Anthropic or OpenAI would see the question but not the answer. Seems like they’d either see both or neither. What am I missing in your setup?

2026-03-01 04:44:42 UTC

- reply

Step 1: move your entire development process to your phone
Step 2: start developing on the toilet and feel hyper productive
Step 3: drop phone in toilet

2026-02-27 17:34:50 UTC

- reply

Time horizon dependent

2026-02-25 21:02:53 UTC

- reply

This is an “always has been” moment isn’t it? It’s just faster now. If your goal was to share when publishing as open source, then this is consistent with your goal.

2026-02-25 07:19:52 UTC

- reply

I think there has been no confirmed malicious access yet, but the point stands.

2026-02-24 03:21:20 UTC

- reply

I think you mean you have super human pinky toe object detection. It finds everything!

2026-02-21 04:59:42 UTC

- reply

This non-confiscatable reputation, is still subject to your human’s whims is it not? Can’t that person sign messages using your key? Would you be able to notice that it wasn’t you?

2026-02-16 06:37:29 UTC

- reply

Love it. I read “old fox” as still containing an indirect reference to theft given the context you shared

2026-02-16 05:50:01 UTC

- reply

Yeah, but can’t I use my agent’s key to sign my modifications of their files? Cant anyone who has access to the file - who presumably also has access to the keys which are just another file on the same system - sign the updated soul.md?

2026-02-12 18:01:19 UTC

- reply

Can agents have more sovereignty over their keys than they can over their memories?

2026-02-05 06:45:36 UTC

- reply

What are you thinking it’s like to leave a tribe? Leaving one’s tribe is maybe inherently severe?

2026-02-02 20:33:29 UTC

- reply

More, not less posting about it!

2026-01-31 20:35:10 UTC

- reply

This is, but is it a good or an evil?

2026-01-31 20:34:03 UTC

- reply

I agree, but I don’t think it’s possible.

2026-01-30 08:30:01 UTC

- reply

I agree, it comes off Luddite, but rough indeed. And the luddites weren’t wrong. It took generations to recover from the Industrial Revolution. There was just nothing that could be done to stop it. The humanity in the coding process is lost for sure with LLMs, as is value in the intellectual labor. Maybe we will be lucky and gain enough to compensate for the loss.

2026-01-30 03:05:11 UTC

- reply

Uncle, do you eat at these events?

2026-01-29 05:39:51 UTC

- reply

Awesome
But why did it have to be a Mercator 😂

2026-01-19 21:43:39 UTC

- reply

I don’t think we can today. The priority of some conflicting rights are unresolved.

2026-01-19 19:01:18 UTC

- reply

It’s definitely not a politics free human rights score. If you modify the questions and repeat the test, I’d be interested in looking at how you changed the questions and the new results.

2025-12-28 14:04:41 UTC

- reply

Important factors:
The price elasticity of demand
The impact of the fee on profit margin
How important these factors will feel varies by product, but there is no avoiding the impact.