>I really don't want to read anything generated by something like Opus 5 at this point.
Personally, I find that its generated prose tends to have an undue weight to it, almost as if every topic I ask about somehow bears a heavy burden, or is otherwise load-bearing, to use its parlance.
I have a personal theory: LLMs are *fundamentally* handicapped at perceiving what's going on in the mind of the human (this can't be "innovated away") and that's at the root of what makes them suck at conversation.
Next time you're chatting with someone, notice how much understanding is shared without anything being said. E.g. the other person might share something deeply disappointing, and they can tell without you even saying anything whether you get what they're going through. This unspoken-yet-communicated information guides the conversation. Or as another example: humans can read the room -- you walk into a room and immediately adjust your demeanor based on what you see and sense.
LLMs are totally blind to things like this, and this adds an inescapable awkwardness to interacting with them. I don't believe they'll ever grow out of this. Which thankfully implies more long term demand for humans instead of robots. :)
Very well observed. I found one more thing: they fail to consider what a 3rd person might understand from your conversation, so when you ask it to dump stuff into a Documentation, they keep making references to facts you had previously discussed or to the train of thought, completely irrelevant to bystander.
100% -- this is the worst. Referencing all sorts of "words with made up contextual/analogous meanings" based on the conversation...outside of the conversation.
Does anyone have a read on if this is primarily a Claude issue, or if all LLMs do this?
I think it's fundamentally and issue with LLMs, for this to not happen they'd have to constantly think "what does other person think right now/what's their state of knowledge" AND also apply that to am additional, third person which would be reading the Docs. They can't even do the first bit well. I think it's a limitation we'll have to live with.
This is basic empathy. It's one of the key differences between a bad and good teacher. People who can see where you're coming from and grasp how your mind is working and correct that train of thought. Meanwhile the bad teacher will simply answer your question. You keep on parroting rather than understanding.
It has a sort of metronomic quality. It never slows down or speeds up or modulates its tone. It plods forward at a relentless pace and never has a light touch with anything.
I think this is one reason why LLM text is pretty exhausting to read for long stretches.
>It has a sort of metronomic quality. It never slows down or speeds up or modulates its tone.
It's possible that this quality you describe stems from the extensive training corpora utilized by the major AI labs. These almost certainly include work from the esteemed economist Jacob Silj:
Not OP but I just watched the episode he's referring to - Season 3, Episode 9, titled "Daily Active Users" where the company reaches a huge number of installs before realizing that the metric that matters is not number of downloads but number of daily active users (actual users as opposed to people checking out the product once) - so he means to say that the number of downloads is less important, scaling matters at sustained use and returning customers I guess?
I lasted about a week before giving up on 4.7 and reverting to 4.6 myself. It introduced so many regressions it was nuts, then failed to troubleshoot the very regressions it introduced, leading to a vicious cycle that tended to compound itself.
Also Wazzup? was later. And the dialog looks different in screenshots you find online but then again there have probably been quite a few releases. Still, I would say those images are not real or maybe from some modded version.
It occurred to me recently that AI's degradation of the human factor via way of increased pressure on the remaining ranks of humans might actually be far more damaging than the AI's output itself.
>Dennis is the best, but the book did him a disservice by painting an unrealistically sunny picture of him as some kind of visionary figure.
Wait, 'unrealistically sunny'? You better not be talking about Dennis from It's Always Sunny in Philadelphia, because we're all screwed if so.
Then again, the western AI landscape has become somewhat stale recently. Claude and Gemini may have cute names, but they all pale in comparison to The Golden God.
>I cannot see any reason, over than oversight and a lack of imagination, why something useful in Ukraine in 2022 was not feasible or useful in 2017 by the USA.
Perhaps it had to do with optics? It's not like there was a lack of capability in 2017. [0]
The war in Ukraine provided a way for the US to assist in rapid iteration of the technology without having to shoulder the negative sentiment or grapple with the morality of it.
Also worth noting that the two conflicts were wildly different: Afghanistan was more of an occupation across a much larger area with air superiority. There's not really much impetus to field killer drone swarms when you already have the 24/7 ability to instantly delete most enemy combatants off the map to begin with.
Whereas Ukraine with neither side having air superiority and it resembling something closer to modern trench warfare. In most cases with literal trenches.
>We already used drones quite handily well before that time frame but in a much more limited manner in a different form factor.
The picture below is from 1995. [1]
By approximately 2001 it received the MQ-1A designation indicating it was capable of employing AGM-114 (hellfire) payloads. Kind of crazy to think about.
_951205-N-3149J-006.jpg){kind=link}
Personally, I find that its generated prose tends to have an undue weight to it, almost as if every topic I ask about somehow bears a heavy burden, or is otherwise load-bearing, to use its parlance.
Quite puzzling, really.
reply