I had exactly the same thought around Apple’s “report spam”. I have marked hundreds of text messages as spam. There is enough training data in there for even a dumb classifier to filter some of this stuff out. But next day, there are half a dozen more, nearly identical spam msgs sitting unread in the inbox.

I have also found that replying STOP = sign of life, your number is marked as “someone alive on the other end” and gets resold to 20 new lists.

Perhaps the real answer is people’s buying power. If people can’t withhold their work, they could withhold their spending, ie boycott misbehaving entities.

>> Yes. It removes political power. As a thought experiment, what happens when sanitation workers go on strike? What would happen if all trash collection was automated?

>> What happens when a group of people goes from the driver's of the economy to a cost center?

> Perhaps the real answer is people’s buying power. If people can’t withhold their work, they could withhold their spending, ie boycott misbehaving entities.

But if they don't have work, they won't have buying power to withhold. Post AI-displacement, the economy will reorient around serving the people who still have buying power (the oligarchs and their hangers-on).

Expect demand for food to plummet and demand for dick-waving pyramids and rocket cars to skyrocket.

I came to this conclusion too and felt rather defeated, existential crisis, etc. But the more I thought about it, the more I realized that if we had a country of 250 million unemployed angry folks (many of them armed), I'm not so sure it'd be the wealthy in control.

This is why there's a huge amount of investment into police technology and drones.

> But the more I thought about it, the more I realized that if we had a country of 250 million unemployed angry folks (many of them armed), I'm not so sure it'd be the wealthy in control.

It depends on how hard they'll fight, but technology just doesn't seem like it's a force for good anymore. I could imagine that, in the face of a revolt against automation, some fraction of the oligarchs will just establish a Ukraine-war like AI-drone kill-zone to protect their assets. If they do that, they'd probably win.

But it probably wouldn't get to that point, because, given the online nature of culture, AI-powered surveillance and AI-powered social interference could very likely thwart any revolt before it starts.

Luna is incredible for the price but I find 6 to be worse than 5.6.

This is a touching sentiment. Thanks for sharing it.

The problem with CSS is the cascade and how it is not scoped. It's like global variables in code. And that can be traced back to it's roots 20+ years ago as a way to style "documents" vs applications that are naturally composed of self-contained UI components.

It should be noted that as of now there are pretty good ways to deal with the cascade

They are better at comparisons between two items vs scoring it in isolation. You can then turn a set of pairwise comparisons into numeric scores.

Write tests first. Have agent iterate until they are satisfied.

The point is that it boils down to writing the tests correctly, regardless of who is implementing the actual code. Hand-written code without test coverage has the same problems as AI generated code.

Ok agentic Ron Jeffries

https://ravimohan.blogspot.com/2007/04/learning-from-sudoku-...

> Have agent iterate until they are satisfied.

Or have them rig the tests so that they always pass.

You don’t even need to prompt them to do that, they’ll do it by themselves. So convenient!

Now that the models are smarter / know more than most of us, a new bottleneck is finding out what you don't know. If you don't know about logistic classifiers and how they could be applied to your problem, you will not ask for it.

These days, I tend to start my coding sessions by the high level problem I'm trying to solve vs the prescriptive, specific solution I may have in mind. It often surfaces ideas and approaches that I did not know about.

I was in a similar situation and tried a different approach.

I had start with asking for a contact with some details I provided. But then I ask the model to be an experienced corporate lawyer and ask me a series of questions to gather the details it needed and then write the contract. The result covered a lot of details that were highly relevant but were absent from the original attempt.

The key insight is that you can lean on the model to cover your unknown unknowns.

Your comment is a great example of the phenomenon where people think AI is an expert in areas that are not because they have no way to determine if the seemingly good looking result is genius, gibberish, or somewhere in between.

The point was that you can go further than "write me a contact that says x,y and z". I actually believe the process I went through with the AI was a more custom, tailored experience than any interaction I have had with actual lawyers, who'd usually apply a template. The AI asked me a bunch of questions, many of which were relevant, and led to follow up questions. I could ask my own questions, make corrections, etc. and all along, they made sense to me. This part of the experience was great.

There could still be more unknowns that I did not encounter and by definition don't know about.

And the final output could be wrong, wording, exact legal terminology etc.

Can we say working with an actual lawyer is guaranteed to be better, more correct? Lawyers are people and they also make mistakes, wake up on the wrong side of the bed, hate their job etc, just like the rest of us.

Here's another techniqe that gets you (even) more out of LLMs: run each prompt several times and compare the various outputs. It is not uncommon for models to contradict their own advice, and also you will get additional insights not included in previous runs.

This is due to the fact that LLMs are statistical processes that rely on pseudo random numbers in chosing what to say and how to say it to a substantial degree.

> The result covered a lot of details that were highly relevant but were absent from the original attempt

Another one suckered by the plausibility engines.

Did you validate the output with a lawyer?

We have been working in this space for the past year. Based on our experience, I no longer trust any claims unless they are backed by benchmark results (yes, benchmarks are painful to run reliably and expensive).

It is possible to reduce token usage. It’s just much harder than the basic approach.

One quick win is to just avoid wasteful tokens, for example run all the QA tools like the unit tests in --quiet mode, which only prints warnings/failures.