Giving someone the text output of a LLM is very similar to publishing a summary without links to the referenced material. When you were querying your LLM, you could have asked specific questions or asked for a custom focus or point of view. Your intended audience might have questions or different concerns, but they're unable to interact with your LLM. What you have delivered is static and unresponsive. It has all the disadvantages of being machine output without the advantage of being interactive, the way your LLM was for you.

It may have to wait until compute is cheap enough that tokens are essentially free, but we need a system to pass "hyperlinks" to LLM's primed with context, ready to be interactively queried on a chosen context. It's being overly generous to assume that people are putting even 300 bits into a LLM for every 1000 bits of regurgitated writing they try to pass off as their own. When people post LLM output as if it were their own, I have no choice but to assume they had zero knowledge of the subject, but this query taught them what they wanted to learn, and now they're sharing that. That's fine, but please pass an interactive LLM link rather than static text.

Once we have "hyperlinks" for LLM sessions, perhaps we can share LLM output a little more usefully and honestly.

Communication only works if you have multiple levels of representation and abstraction, including but not limited to - letter shapes, grammatical structures, style and register, stylometry, and subtext.

All of that is learned, and writers usually assume they can rely on that learning as the context for the text.

So you don't write to 'transfer information' like a network cable, you write to trigger experiences in the human version of latent space.

Factual information is one kind of experience. But even when that's the goal, there are always layers of implied relationship, social register, role, status, and other implications in everything that's written.

In normal communications the context - business emails, personal messages, mainstream journalism, fiction, and the rest - defines what acceptable language looks like.

The content fits inside that. But it has to fit the context, otherwise it lands in a semantic and psychological uncanny valley - like sending LinkedIn speak to a spouse on a wedding anniversary.

The real problem with LLM writing is that it's good at the technical layer - the grammar and spelling - and has some insights into the rest.

But the default content style is marketing and ad speak. And recently it's developed a weird and unique hybrid style which applies marketing fluff and pretension to technical content like code comments.

So you get one register instead of all of them. It can attempt others, but it's still too limited to generate them fluently. Sometimes the results are outstanding, but often it defaults to mechanical clichés.

So that's why it sucks and sounds so hollow.

Can it be fixed? Yes, but it's very hard work, most people don't have the skills, and it takes time - often too much time to be worth the effort.

I see this all the time now with LLM generated output. It's easy to have an LLM generate a chunk of content that can be dropped into a chat or comment, and when it took you 20 seconds to have something written up based on the shared understanding you and an LLM have about the context of the situation, but it takes other people 3-5 minutes to read and understand that content, that fundamentally doesn't scale. It's bad enough when one or two people are doing it, but if the whole team is doing it, the only way to keep up with the stream of information is to also consume it through an LLM. At that point you're likely to be missing much of the nuance, and the amount of errors will explode.

This can be alleviated by people reviewing the output of an LLM and making sure it both includes fundamental information that might be assumed by context and reducing it to the parts that are essential for the new context it's in. This takes time, but is extremely important.

Having an LLM write gobs of text to send to other people instead of doing it yourself is the equivalent of a low yield cognitive zip-bomb. Don't do it.

No, I don’t want to read LLM writing because it is BAD at it. It doesn’t really understand how humans think (because it thinks differently), and doesn’t seem to understand core principles very well (presumably due to the lack of world model), so it can’t write something humans enjoy yet.

https://github.com/josephmisiti/awesome-machine-learning

It's helped a lot. Agents haven't figured out how to do that yet, or sendgrid, sns, etc are doing the hard work for me.

In the meantime, for business communication, I use AI to shorten my text, to make it more concise.

So far I haven't had a reason to go back through commits to isolate any issues but if I do hoping the 'why' messages may come in handy for my LLM lol

I've only just discovered pangram, but I've seen it referred to a few times in HN recently wit nothing obviously pejorative about it. Take it with all required grains of salt, though

0: https://www.pangram.com/dashboard

If this actually works, it's absurd. The implication is that Anthropic could trivially make Claude sound less obnoxious, but chooses not to. I don't see a way this could be justified as a safety feature or anything, so.

I agree, using LLMs is disproportionately frowned upon, but what matters is if you invested your own attention in the process. I use LLMs a lot for sparing, usually ask it to assume some opposing persona or use web search, not relying on its defaults.

Everyone is an elitist, when you've enplaned do you assume the pilot is qualified to fly it or do you hope everyone gets a turn? Admittance to the canon is no less a responsibility, would you trust the yoke of human culture to the artless, craftless, and naive?

not everyone does their thinking, if you dont I dont want you to have voice

I still want the imperfect human.

As for second paragraph, the writing "had no overall meaning". There is nothing to respond to if there is no overall meaning.