It’s a bit of a semantic grey space, like many words. For common use, anarchy and chaos are synonyms, hence why your initial comment could be read both ways. For a certain class of ‘rebellious’ individual, it’s used more like a naive, ‘lower case l’ libertarianism. For some, it means the absence of any social structure at all, a ‘state of nature.’ For some others it’s the de facto reality of all systems using a definition of ‘who has the most capacity for violence makes the rules.’ For those studying sociology and anthropology, it’s used specifically for a class of societal organizational systems that may be highly organized but share a lack of hierarchy. The shared element between the various uses is the lack of structure so I lean toward keeping it to that basic concept and hesitate to claim any of them are the ‘correct’ definition.

Why…?

Because why bother saying anything if you aren’t going to say anything? Offering correct information gives the other person a chance to correct and improve. Just saying ‘WRONG!’ is just a slap in the face that only serves to let you feel superior, masturbatory pretense.

As for the rest, those are all clearly issues, but none of them are of a sort where handling the one I raised and handling them are mutually exclusive. And at least the second item is actually a following point from the one I mentioned. People being tricked into thinking LLMs are capable of thought contributes to the thought by decision-makers that people can simply be replaced. Viewing the systems as intelligent is a big part of what makes people trust them enough to blindly accept biases in the results. Ideally, I’d say AI should be kept purely in the realm of research until it’s developed enough for isolated use as a tool but good luck getting that to happen. Post hoc adjustments are probably the best we can hope for and my little suggestion is a fun way to at least try to mitigate some of the effects. It’s certainly more reasonably likely to address some element of the issues than just saying ‘WRONG!’

The fun part is, while the issues you mentioned all have the possibility of creating broad, hard to define harm if left unchecked, there are already examples of direct harm coming from people treating LLM outputs as meaningful.

They’re both technically anarchic, (no hierarchy among rioters either) but things like this demonstrate the lack of hierarchy is clearly not the problem in either situation.

I keep wondering what the world would be like if the prices of everything were not subsidized by exploitation. So many industrial processes are done in places where they aren’t made to adhere to HSE regulations because it’s cheaper. How much would your day-to-day expenses increase if it meant all the bits and pieces involved cost enough to pay for vaguely safe and sane working conditions and environmental protections?

Good. I couldn’t be sure. ‘This is actual anarchy’ is just as readable as ‘this is the degeneracy of our modern culture’ as it is as ‘this is people acting responsibly without need of hierarchy.’

What do you think the most important issues with AI are? I see a lot of ‘you’re wrong’ but no indication as to how or why.

At the purely GUI level, if you’re being granted acrobat, it turns out you can extract arbitrary subsets of pages manually, very quickly. You can then rename them. I haven’t learned powershell personally but it absolutely could be used to batch rename files, even if it’s a somewhat silly looking language compared to bash. Again, though, how much work that involves depends on your desired naming conventions.

It could be if someone has stress-induced impotence but then learns coping techniques and it goes away.

Very corporate thinking. ‘We have the IP for the name of a game with a dedicated following. That’s more important than who makes the game, right?’

It has gotten so much easier too. I started with Ubuntu on an old laptop to try it out back in ~2017. I had some ‘learning experiences.’ This year, it took almost no effort to slap bazzite on a machine, do a few small tweaks using a GUI settings app, no terminal activity needed, and let even someone who has never used a CLI just get on with their day with basically no issue. At this point, the only issue is a handful of software, specifically Adobe and their nonsense.

The magical IT field is emitted in a cardioid aligned to the forward axis of the generator. Its effect is inversely exponentially proportional to the distance from the generator. This is why almost all IT problems are immediately solved only when you have put down whatever you were working on and actually started going over to help.

No, it’s pretty much the opposite. As it stands, one of the biggest problems with ‘AI’ is when people perceive it as an entity saying something that has meaning. The phrasing of LLMs output as ‘I think…’ or ‘I am…’ makes it easier for people to assign meaning to the semi-random outputs because it suggests there is an individual whose thoughts are being verbalized. It’s part of the trick the AI bros are pulling to have that framing. Making the outputs harder to give the pretense of being sentient, I suspect, would make it less likely to be harmful to people who engage with it in a naive manner.

Hmm. Well, first off, if you mean you don’t know how to write a script and don’t view it as worth learning for this task, that limits the task a fair amount. If you mean you don’t want to learn about the particulars of script based PDF editing or OCR, that’s understandable.

If you don’t want to script at all, you should be able to segment the PDFs via acrobat, or even just ‘print to PDF’ with page ranges on most viewers. There are ways of bulk renaming files once you have segmented them, even without scripting, though it’d be use case dependent as to whether/how that’d be useful to you.

If you want to script just a little, I made a script ages ago where I used the documents’ name to hold the metadata of what needed to be modified. You could certainly do that. (e.g. open the doc in one window, select the file for renaming in your file explorer, scroll through and input the sequence of pages in the rename field, [documentName3,7,15,22,29.PDF] run a script to segment the PDF at those page numbers so you end up with ‘documentName-1.PDF’ containing pages 1 to 2, another with 3 to 6, etc.)

A bit more effort could maybe be used to do some level of renaming, though how much use that would be would depend on the particulars of your case. I could see extending the previous script a little and making the page annotations include a doc type. (e.g. 13cn meaning segment at page 13 and label it as ‘originalDocumentName-clientNotification’, or even 13’arbitraryText’ and use the arbitrary text as the new file name)

The particularity of your case may be precisely why it hasn’t been automated yet.

I know there are scripting ways to work with PDFs. I was listening to someone talking just earlier about using a script and a localhosted LLM to organise and rename PDFs with author and title. If you can identify some kind of patterns (such as a heading that starts each document of a type) that you can detect, a script could find those pages and then feed that into something that will segment page ranges for each doc. It’s definitely possible but the patterns to look for will be determined by the docs you are looking at.

Possibly.

Hmm.

Short answer: potatoes.

Famously, you can survive on a diet of only potatoes. Starch gives you the caloric energy. Skins give you nutrients. Potatoes can be bought in large amounts for not much money in many places.

Everyday people coming to the aide of a fellow human who is fighting to stop herself being victimized is anarchy? Yeah, pretty much. Doesn’t seem like a bad thing, though. If we all took that level of responsibility, you wouldn’t need much of a governmental force.

There’s a lot of ink spilled on ‘AI safety’ but I think the most basic regulation that could be implemented is that no model is allowed to output the word “I” and if it does, the model designer owes their local government the equivalent of the median annual income for each violation. There is no ‘I’ for an LLM.

People are very aware of those. Many have them, and many of those who have them are all too willing to share them.