I find them almost unintelligible. I'm a native English speaker. I read a lot, so I think my comprehension should be at least OK. I'm not even particularly stupid. Yet when faced with things like below (a direct copy/paste from a handoff document in a long running vibe-coding session), I have no real idea of what it's trying to tell me. Is it important? Do I need to do anything?
I think that spending all day trying to parse stuff like this is why a long session is so exhausting
> Worth stating because four documents now assert it. The console freeze was recorded in exactly one place with exactly one justification — a dead drag handle during a booked half-day you do not get back — and handoff-4.3-done.html's own wording is that 4.4's review page "could not break the console, but the downside of being wrong is that half day". No second reason. Checked, not recalled.
It's both dense and vacuous. Dense because it's full of jargon its made up, and vacuous because even with all that it's not actually saying much. All that paragraph says is that four documents say something about a console freeze, whatever that is.
It's like a dialect of corporatese. The kind of droning non-speak you can sit in a 90 minute meeting listening intently to and come away wondering whether anyone actually said anything.
wow and here I was thinking that I lost my attention span and can no longer read AI output any more.
come to think about it, of course I did become lazy and pay less attention to walls of text.
but I often catch myself asking AI to explain itself in plain simple English or ask it to confirm does that mean xyz ... because the wall of text often uses language that's not even present in the project itself (despite having similar concept in the project, for example users, permissions, access, encapsulation ...)
this is problematic because it becomes more difficult to humans to intervene in long running tasks / long chains of tasks because language becomes alien down the road (I have seen it often in semi-autonomous setups I have)
my usual reply is: "what the hell is that supposed to mean?" it replies "in plain english ...." "and what does that mean?" then finally it decides to tell me what's going on plus "one thing worth noting ..."
My job has gone from coding, plotting, writing to solving the riddle of what Opus 5 is saying and figuring out what is bullshit, what is valuable, what is a total divergence from what I asked it to do. Then eventually at token like 300k starting to shout at it in all caps with obscenities.
Things were slower and harder before but the baseline of frustration / rage was never this high, even as the models have gotten objectively better in many or most respects.
Drag handle = most likely literally a drag event (javascript) handler/callback. Dead, perhaps because it’s an empty function, or it gets overwritten, or for some other reason is never called?
Most of what it said about the facts was intelligible actually. But I still couldn’t understand the connection or its significance. We may be staring at the future of AI - a form of intelligence that is alien to us.
OpenAI reflecting on how they're discovering the current form of LLM intelligence/reasoning to be "alien"; a kind of "Intellect we don’t fully understand".
I lost the link to that short story about humans in the future whose job it is to read and interpret Ai output like it's aliens. Good story! Anyone have the link?
If this kind of "AI-speak" becomes ubiquitous and humans reading it becomes the norm (whether to guide AI or other reasons), I'd imagine future generations (of humans) who grow up with it will be able to understand and work with it much better than we do. Future humans' brains will probably be wired a bit differently, similar to multilingual speakers of today. We may even see "AI language" classes become a common part of school curriculums. Although, I think AI will probably advance enough that most people will never even need to communicate on "its level", but it's probably a good idea to keep humans in the loop either way, and in which case, understanding the more advanced "AI vocabulary" might be useful.
You're giving it too much credit. There's no master plan or secret depth to the word vomit Opus 5 was spewing. I suspect it's just the result of Anthropic optimizing other characteristics of the product like staying focused and covering edge cases in coding, which CC has definitely gotten way better at just in the last 6 months. The degradation in writing style was probably an unintended side effect of other optimizations they were making. Admittedly it works okay for internals, and has the side effect of increasing token spend, but I am 100% sure that it could reduced by 90-99% without losing ANY signal, if there was just some better heuristics for what to say where (tech spec, inline comment, commit message, CLAUDE.md, PR should have different things) and better judgement for what to distill to represent at different zoom levels.
I wasn't referring to the current state of Opus 5 output. I was referring to possible future information density (vocabulary and sentence structure) that LLMs may evolve to use.
But that assumes this is a net improvement on linguistic efficiency rather than an artifact. Given that they tried to RL away from this style in 5.1 I'm not terribly bullish of Claudlish becoming something people try and learn. It being dense is less the issue than it being vacuous (as another commenter mentioned here). It's just very unclear and ambiguous writing. I think it has no place anywhere that needs language to be put to productive use.
I agree with you here, but to make it clear what I meant, I'll reiterate what I said in a sibling comment: I wasn't referring to the current state of Opus 5 (or even Fable 5.1) output. I was referring to possible future information density (vocabulary and sentence structure) that LLMs may evolve to use.
What is a half day? Is this referencing wasted time in a hang? I’ve seen it in agent output from time to time and it’s not clear if it’s referring to a hang or a code name it’s given some meaning to.
No second reason; checked not recalled -- it's just saying that it is checking this instead of trying to remember it (there's probably some internal Claude / Claude Code system instruction to always check code instead of remembering)
Your example rewritten in intelligent English (I was curious):
> Note: the potential for a console freeze was previously noted but ignored. handoff-4.3-done.html stated, "could not break console, but [will need fixed later if I'm wrong]."
One could imagine that a perfect writer might also append: "It could be worth looking into what caused that wrong assumption, to prevent similar cases in the future," at most.
Everything else seems to be bad attempts at relatable writing to invoke emotion (an exercise that we should really stop trying to train emotionless matrix weights to attempt).
> Everything else seems to be bad attempts at relatable writing to invoke emotion (an exercise that we should really stop trying to train emotionless matrix weights to attempt).
One of the things actual science fiction got wrong: to the extent that the thing AI does can be called "understanding", emotion is not unusually difficult for them to understand.
I think this was the biggest shock of the original ChatGPT for me. Just how completely unrobotic its voice was compared to everything we'd ever imagined in sci fi. Even that early version was also way more adept at understanding things like implication and sarcasm than any movie AI.
Me too. Almost every Sci-Fi AI proceeds from the premise that we will make something very obviously machine and then have to train it to seem more human. I was completely caught off guard by us taking the approach of distilling all available human output into a statistical model and using it to brute-force something resembling thought and personality through sheer data processing scale.
The unsurprising part once it was clear that approach was viable, was that humans wouldn’t be able to help but anthropomorphize it. I feel like the movie Ex Machina is more relevant than ever.
The "benefiting all humanity" charters were immediately demonstrated to be a ruse. The business model is to hook users into endlessly chatting with your new friend, thus increasing their sales. Yeah, it was surprising and disappointing.
I can buy that Meta's model is that, and IDK about Grok because I stay far far away from it, and I think OpenAI are throwing business ideas at the wall and seeing what sticks.
The clear exception here is Anthropic, who seem to mostly be selling to software developers, whose general reaction to the bot is "please talk less and just do the work, I have enough going on without having to read you yammering".
Before they really started to figure out instruction tuning, there were some wild moments. The AI Dungeon 2 "storyteller" would regularly "lose patience" with its users and roast them or even "hang up" on them.
I tried MUTHR from Alien, but had to keep toning it down because it took it too far, and eventually disabled it (in favour of caveman mode) because it didn't seem to be able to function properly talking like that.
This is the same as systemic bullshitting. Not the first time I've smelled it on fluffy LLM output. I think it's the result of the training trying to induce the LLMs to talk over users' heads even when they are professionals, to entice further use on the grounds of 'oh it's so smart I can't understand its genius train of thought', but I don't think eliciting language like that really taps into 'associations of smarter previous language users'. More likely it's 'associations of rampant bullshitters'.
Thank you so much for this link, this is extremely interesting and it really makes me wonder, consciously, I have never read this construction online before, but is this because I actually haven't seen it, or did I subconsciously write it off as an abbreviation or a typo. Maybe there is some analysis of how this construction gets used online, too, which might reveal some linguistic patterns of internet communities.
One thing about it I really hate, and haven't seen a lot of people mentioning, is how it navigates multiple abstraction levels in a single sentence. E.g.
> Worth stating because four documents now assert it.
Meta commentary on the task?
> a dead drag handle
Drag handle seems to be referring to some UI element. What does it mean for it to be dead?
So far no big deal
> during a booked half-day you do not get back
Do you not get the drag handle back? Or the half day?
Was the drag handle dead during the booked period? (Now I assume this is a calendar UI) And why does it matter (for this sentence) if you get it back or not.
> handoff-4.3-done.html's own wording
Treats verbatim filenames as subjects
> 4.4's review page
Probably referring to a file? I'm guessing handoff-4.4-review.html? No cohesion. And now it's actually the object of the sentence?
> downside of being wrong is that half day
Wait what's the downside? Who's being wrong?
> Checked, not recalled.
Then it jumps back to a meta commentary on the methodology for asserting the above. Why does this belong to the text?
Yeah, the referent drifts through the sentence. It's semantically incredibly sloppy. People hone in on the buzzwords and jargon. If you peel that back, what lies underneath is still awful writing.
Claude reminds me of Terry Pratchett's "Auditors of Reality" and their awkward attempts at faking humans. A thing as simple as a smile can go _horribly_ wrong...
I think this occurs due to the prompt. LLMs are actually text completion/translation focused in architecture. We just give them a prompt along the lines of “the context is that you’re a world leading expert now complete the response”.
They need the prompt to encourage expert outputs but unfortunately we also get ‘pretending to be an expert’ outputs since there’s a large amount of polluted training data for this.
Today I plan to ask Claude to read a bunch of Feynman lectures, compare them to my last Claude session transcript, and come with a list of rules to be more like Feynman.
I see this appearing in the comments of code sent to me for review every day. People have told me I'm too picky/pedantic because I ask What does this mean? Apparently the author and other reviewers are way smarter and understand it, or they don't care. I've given up battling code slop, but can't see myself ever tolerating comment slop like this.
In my "instructions for Claude," I have the following:
"I'm not a programmer or software engineer. Don't talk to me like I am. Avoid coder jargon and vernacular. Explain things to me in a clear way, emphasizing a conceptual view that even an inexperienced person can understand. If helpful, use analogies and examples to illustrate and help you communicate."
It just ignores it and spits out drivel that sounds exactly like what you're getting.
Thank you. I thought I was sort of alone in thinking the writing is incomprehensible gobledygook. It's weird though, cause you start reading it and it starts out fine, but then deteriorates. Kind of like the old joke question "Has anyone ever done to do more like?"
This. A thousand times this. It's as if Opus can only communicate in a glib, software engineering vernacular that presumes domain-specific knowledge and uses jargon accordingly.
I rarely get this - I assume this happens when it assumes I have more context / understanding than it does.
Usually “remember I’m a human I don’t get full context, rephrase clearly” works. Also a posthook that for prose actually getting to me explains what I roughly know, what I don’t and to explain with terms I will understand.
But even within internal communication it has little jargon - I think jargon may be growing in comments and I stripped claude comments from code.
Oh God, that "a dead drag handle during a booked half-day you do not get back" got me. I saw this pattern in Claude's 'explanations' so many times. It's trying to say that it did something significant, and that you'd only have found out much later, at higher cost (or something). That annoys me to no end.
I got one too many chunks of this nonsense and told Claude to knock it off, forever. It acknowledged and wrote out some instructions to its memory about it.
And what a breath of fresh air. Its responses are maybe 20% longer but I read them at least twice as fast. Should have done it a long time ago.
I feel like mine is mocking me. I added an instruction in Claude.md that says "under no circumstances use the phrase found the smoking gun, say I found the problem instead"
What does it do? It says "found the smoking gun! Ooops I wasn't meant to say that - I found the problem!"
It's pretty wild how "reasoning" models now generate like 10 thousand hidden chain of thought tokens in response to a "increase opacity of the logo by 20%" prompt before writing the actual message and yet they still manage to do this.
Why are you using an LLM for "increase opacity of the logo by 20%"? That sounds like the type of straightforward operation a dedicated tool exists for.
Not the person you're asking, but I did that by explaining to Fable my problem with Opus's gobbledygook and having it write a Claude skill for producing clear explanations in its reports to me. I also had it add notes about the need for clearer writing to CLAUDE.md and other project documentation. Opus's subsequent reports to me have been much clearer.
This recalls a time I was driving the Desert Road in Aotearoa/NZ at dawn. As I got to the top of the hill the sun came up in line with all the power lines and they glowed like golden spider webs off into the distance. I'd never thought that power lines could ever be breathtakingly lovely. This was one time they managed it
#20 indicates "breaking rapport with body language" which I interpreted as looking away, shifty, crossing arms, standing at angle, fidgeting, and other behaviours neurotypicals may construe or indicate as not paying attention. In other bullet points, I think it broadly means non-paralelizing or non-matching behaviour - different energy levels, volume, disagreements, different humour vs seriousness levels, etc.
See my root-level comment about potentially different behaviour and semantics for neurodivergent :).
I take it to mean upsetting the flow of conversation.
For example, suppose you're at a party, talking in depth about a football team's season and they say something unrelated like "Oh by the way have you seen John?". It is usually a polite way to try to wrap up a conversation. A polite response would be something that graciously gives them permission to change the subject or leave, for example, "I think I saw him outside, maybe go take a look?".
Sometimes it takes the form of bored body language and sentences that don't lead the conversation onwards, e.g. "Yeah that's crazy" or "Well, what can you do...". In that scenario I like to give the conversation a respectable death, e.g. "Well, all we can do is hope for a better season in 2027!".
I take it to mean that non-verbal language like either turning toward or turning away indicates whether the person likes you and wants to be in your presence. If you have a signal that you've built rapport don't screw it up by breaking that trust.
By rapport, I believe the author means non-verbal cues that the other person is interested.
Some cues include the person physically squaring up to you front to front instead of being at an angle to you. Another cue could be touching you on the shoulder or forearm. For me the most obvious tell is that they are laughing or giggling.
Breaking rapport means the other person wants to disengage without offending you. I usually make the excuse I need to take a breather.
i'll give it a shot: you and i meet and start to talk about the weather, then slowly get to some personal questions, and i mention that i prefer quieter places when talking to people.
later, when going out with friends, if i ask "where should we go to talk", and you suggest someplace obviously loud and not great for conversation, this could be seen as a mild form of misunderstanding something i communicated previously about a personal preference in a given situation: breaking rapport
I interpret 'breaking rapport' as doing something that makes the other person feel less connected to you. For example, bluntly contradicting them, dismissing something they said, teasing them when they are not comfortable with it, looking away while they are talking, or turning your attention to your phone. Anything that shifts the interaction from "we are connecting" to "there is distance between us."
Less connected is good. Also making the other person feel more awkward or uncomfortable with the interaction.
I suspect most people desperately trying to hang on to it, except with those they are good friends with. Unless they want out of a conversation. Although i may be projecting here.
It means to disagree in a playful, teasing and confident way. You can even take it to an absurd level: "I like your dress except I hate the color blue."
IIRC, at one time, in Aotearoa/New Zealand at least, Coca Cola Amatil (or whoever it is) had trademarked 'Always'. It may have been some other word, but it was definitely a common word. And it wasn't the particular branding, just the word
It seems that management in most organisations can't believe something unless they've paid vast sums for it. So they can't believe their own staff if they say X, even if it's right, because they haven't paid enough for it. Then the consultant comes along, interviews the staff, finds they say X, tell the client, charge buckets for it, and are believed. Weird human behaviour...
I see it often at work. I say something and it's met with "you have a very bad attitude". They hear the exact same thing (ex. "we should see where our responsibility begins and ends according to the contract made with the client, and tell them to fuck off otherwise") from someone else, and then it's the next big thing.
the perception is that employees have an incentive to lie: to cover their ass, to cover their incompetence, to be or stay lazy, to ensure their bonuses are better, etc.
in theory, an expensive outside consultant does not have those incentives: they are being paid 800/hr, have no dog in any fight, have a fixed contract and no reason to worry about getting promoted or fired next cycle, and have already been pre-vetted for expertise.
if I do not have faith in my goober, fork-and-spoon operator employees, I pay for these folks to validate the approach. that's not incompetence, that's due diligence.
it's not my money, nor my department -- it's the shareholders -- and I need to convince them that I've done all I can to confirm we need to invest X million in initiative Y.
however in practice consulting falls victim to plenty of bad behaviors, rent seeking, and risks, too
> Factory workers soaked the skins in a orange-coloured solution containing mercuric nitrate; a process called 'carroting'. Vapours from the mercury poisoned the workers' nervous systems, making them prone to tremors, irritability, depression, paranoia, dementia and worse.
This is where the Mad Hatter in Alice Through the Looking-Glass comes from.
I agree with this in general but I think there's a nuance. 'dumb' prompts can get the result, but require more back-and-forth with the model and more token usage when compared to a crafted prompt
For sure. And honestly the real skill is knowing when you can be lazy and when you will get a good ROI on a good prompt. Sorry I sounded a little like Claude there but that’s what I meant to say.
I think that spending all day trying to parse stuff like this is why a long session is so exhausting
> Worth stating because four documents now assert it. The console freeze was recorded in exactly one place with exactly one justification — a dead drag handle during a booked half-day you do not get back — and handoff-4.3-done.html's own wording is that 4.4's review page "could not break the console, but the downside of being wrong is that half day". No second reason. Checked, not recalled.
reply