galaxyLogic

Game of Life comes to mind: Most simple logic, emerging patterns are hard to believe.

Isn't the LLM simply predicting what should be the next sentences after user's input, using its algorithm and data it has exatrcted from existing texts on the internet. The algorithm that does that could have many different designs, some better some worse for the purpose of predicting what output makes most sense next?

So what is it that we don't understand about why theyr work? The algorithm? We have the code. Why the specific algorithm makes such good predictions? I see it as a generalization of trying to predict who wins Kentucky Derby.

LLM is an Oracle

The big breakthrough is we can interact with the agents using natural language - because of the LLM.

It is the combination of LLM and agent-harnesses that make it look really smart. Agent-harness is a programmatic device that lets us tap into the vast knowledge in the LLM.

It is probabaly true that many TV-commentators fail to appreciate this fact and therefore think LLMs are super-intelligent. No, it is the combination of LLM and the programmatic agent-haness that is the breakthrough.

An interesting thought is that the LLM could in theory code the agent-harrness, start it running every time we interact with it. Currently the agent-harrness I think is pretty static I think. In theory it could be dynamically created for every task. Would that make it better don't know.

But more thermal storage you want more salt you want, and it's gotta cost something, right?

https://en.wikipedia.org/wiki/Molten-salt_battery

Not much. I worked for an E-Learning company some years back when it was the trendy thing. So I started thinking are the courses we provide really helping students to learn? But turns out the business-model was mostly to provide compliance courses so that companies could prove yes they did train their employees in compliance so they could not be sued. The courses were made kinda easy to begin with because what was important was that employees could pass them without wasting too much paid work-time.

I also gave some presentations in IT-related conferences. People in conferences often get tired. So once as I entered the lectern I turned on a piece of music I had on my thumb-drive. It was the loudest hardest Rock'n'Roll I had. The audtorium had good loud-speakers. I think it woke the audience up. But soon I saw them nodding again. :-)

I think I read somewhere that salt can be used as energy storage medium? So we could get both water and batteries for renewal energy.

I'm thinking of more simple cases, not like going to court for some reason, but make contracts and file all legal paperwork required. Ensure the company is compliant with whatever laws there are.

Definitetly, except people to which the term-limits would apply really don't like term-limits. And they are in power.

But I don't want to souond too pessimistic, there are lots of good people and progress is inevitable. AI gets put down a lot but I think it will make people much smarter.

Thing about it is the students should be given an explanations about why each topic is important for them to learn to be able to learn more advanced topics.

Maybe briefly show how that adavanced topic will be taught and let them realize they can not possible even start to understand advanced topic because they are missing the more elementary pieces.

Similarly why they can't got further without doing their homework. How mastering the homework exercises let's you solve more problems.

I know that is not easy, the teacher may not quite understand how topics relate, why each of them is needed in a specific order, if they have not thought about that much.

What we usually think as consciousness is our ability to think about what we are thinking. We are conscious about our own thoughts, can remember that sensation and those thoughts later.

But LLM has similar capability, it can look at its previous outputs, and think what further thoughts the it should generate next based on those. It can keep some of its outputs to itself, thinking in its own head,and can examine it previous private thoughts easily.

Not sure how much current LLMs do that but clearly they can at least in theory use their own outputs as their inputs too.

Current LLMs however are not conscious of pleasure and pain which really is the root of goal-oriented behavior. But maybe something like that could be programmed into them.

How would you cause an LLM pain? Or pleasure?

I was just thinking about it, rich people have money to spend in elections to get their representatives elected. When they do they make laws, or remove existing laws, to help rich get richer, so they have more money to spend in the next election. And so it goes.

Solution might be legislation that puts limits on how much money each person can spend on elections. But it may be too late, there are so many rich people in the congress that such laws can not pass.

The rich not only want to get richer they also want the lower classes to get poorer so they will work for less and will have to work longer hours so they will have less time and money to educate themselves, and thus will remain clueless about what is going on.

I haven't tried that but so are you saying I could basically code in JavaScript and then ask Claude to turn it into TypeScript?

I like the strutural typing as well. But I hesitate to use TypeScript because AI tells me this:

It Catches: Mismatched function arguments, missing object properties, and typos in variable names.

It Misses: Invalid JSON from an API, unexpected database outputs, and bad user input.

The benefit is not only about "documenting the contracts" but documenting the contracts in a way that we can trust those contracts can not be violated when the program is running.

That is a very good thing to help us reason about the program, we have invariants we know must hold true if the program does not stop in a type-error.

As I understand it TypeScript does not enforce types at runtime. Am I correct? If so that would signify to me it is not a "typed language", like say Java for instance. Types in TypeScript are more like "annotations" for docujmenting the program. Am I correct?

It's probabaly a good things that Uber-developers are now forced to do some coding on their own. Only use AI where it absolutely helps

Right. That will reduce workload for the lawyers. But will their fees then go down? I'm kinda worried that if I don't give them the LLM produced legal docs for review they will just use the LLM themselves and then charge me for the work the LLM did :-)

It's bit like with doctors, you'll want a second opinion, if you can afford it.

Are you a lawyer yourself?

Good point. Same probably applies to code as well, coders much tell us why they write the cde the way they did. And if they have comments in their code, those are highly untrustworthy because noboy fixes comments if the code works.

I'm going to need some legal help for my startup. But I can't pay much. So I figured I will ask AI all relevant questions, as well as forms filled etc. Perhaps even create a patent-application for me.

THEN I find a human lawyer and give AI's answers to them and say "Can you find any errors in this? Can you improve it?" .

That way I think my legal bills should be smaller because the AI has already done most of the work. What do you think? Which LLM is best for legal work?

The converse is that if you have many unused variables some of them are used in some contexts but not others and then when you use FIND to find occurrences of a variable you will find many which are not relevant and some which are.

Overall the code gets smaller and easier to understand when it only has things that need to be there. If you comment out a an unused variable you can see from find-results that in effect that occurrence cannot matter.

Unused variables are "noise" that hide actually important things.

Not everybody can create and maintain their own applications even with the help of AI. And that would be wasteful because app is not made to serve a specific user but to accomplish specific tasks.

Therefore we will need people who are skilled at creating apps with the help of AI. AI can not do that for us, someone is needed to figure out what apps we need. So AI will empower programmers, not replace them. Of course if you are unwilling to learn how to use AI as part of your tool-chain you might lose your job.

Why not just comment out variables you "want to keep around" but are not used anywhere?

Good point with some irony. Thye don't want to do a better job they want to do an easier job. But a company like E&Y should realize shortcuts like these don't work. And their customers are paying them.

Looks very nice but still takes some effort to undertsand how to use it. Isn't it simpler to ask the AI to create the PDFs for you?

Another part of the new price of oil and Trump's tariffs earlier is that they are a regressive tax. Poor people have to pay more in terms of percentage of their income now.

Trump often said "Tariffs is the most beautiful word". I think he said that because he thought his followers were too stupid to realize that he was transferring tax-burden from the rich to the middle-class and poor. That's why he thought they were "beautiful". They enabled him to gaslight people.

American people are not stupid but many of us have to work hard so there's not a lot spare time nor energy to try understand what is actually happening. Fox News is misleading us purposefully playing emotional tricks on patriotic Americanss.

Now import-companies should get back the tariffs they paid. But does it mean they will lower their prices? Probably not.

It was always the Art of the BS, part of which is always to try to hide the fact that it is BS. But people fall for it and his minions often at least initially benefit from the same BS.

Just because somebody tells us they are not lying does not mean they are not. But some people believe it because it makes them feel great again.

I don't quite get it why they can't take another LLM and vet the output of the first with the second one. Surely they would not have the same hallucinations and would be able to detect hallucinations of the earlier LLM. Maybe it would cost too much in terms of tokens?

I don't know but I would expect it to be relatively easy for an LLM to detect "hallucinations".

I don't quite get it why they can't take another LLM and vet the output of the first with the second one. Surely they would not have the same hallucinations and would be able to detect hallucinations of the earlier LLM. Maybe it would cost too much in terms of tokens?

I don't know but I would expect it to be realtively easy for an LLM to detect "hallucinations".