Speleology aside, just a few days ago in one of HN discussions I was told LLM-produced content is already indistinguishable from human-made. But I'm sure I would have pretty good chances to correctly guess even the exact brand which helped to write this sloppy text. And it's the same with many stories I read.
Stake the claim then. I’ve also read a fair bit of LLM text and this didn’t trip anything for me.
Edit: after reading the author bio and the other articles, I agree with this assessment. Though I will admit, this particular article didn’t jump out as AI to me.
I know what you mean. Flowery and fluffy. I don't see any clear giveaways, but it feels like it and is a bit annoying to read because the information in there isn't very dense.
That falls into the general category of 'authority groping' (yes, I coin my own terms) where a prediction engine without real reasoning, must apply numerous tricks to convey confidence, authority and substance where there may be less or none. Similar to the constant negative-propositions or "not" statement, where usually one to three things that do not apply are articulated, which has the result of credibility accumulation, I suppose, by showing "hey, I am right about not just one thing, but four!" hence the saturated styles we tend to see.
Just because the cave is sealed now, does not mean it has been sealed for millions of years. Even for geology, 5.5M years is a long time. I don't understand how they can claim it never had connections to the outside, just because it doesn't have connections now.
"Combined with the geology and the peculiar chemistry of the cave’s water and air, the evidence pointed to a system that had been closed off since roughly five and a half million years ago."
That quote, from the AI generated article, simply asserts the cave was sealed, it doesn't explain how the cave came to be sealed, nor how it is that modern humans infer the cave has been sealed and for how long.
Geology is most probably largely how the sealed claim has been made .. it was open to the world when the limestone was forming, formed (Speleogenesis) by sky water (rain) permeating down and eating out a limestone void, and "sealed" by thick and impermeable layers of clays and loess (windborne sediment) some 2.5 million years ago during the Quaternary.
There are also genetic drift estimates, which put the "last" outside species to enter at about that time (
One of the most recent animals recorded is the cave's only species of snail, Heleobia dobrogica, which has inhabited the cave for slightly more than 2 million years.
You should look up how many live armed nuclear bombs have been accidentally dropped on mainland US soil.
More significantly, there's been a lot of above ground and atmospheric tests that have draped residuals across the planet, and a bunch of dirty Russian / former USSR sites that drift trace across Europe.
Indeed. That's one of the reasons why the "Before Present" time scale (that uses 1/1/1950 as the epoch) is sometimes called "Before Physics" (another one being that 1950 is less and less "present", of course)!
But also, finding something from within the last hundred years doesn’t say an awful lot about something claimed to have been a closed system for millions of years.
Looking for Chernobyl signature in cave during the late 1980s / early 1990s does seem daft - far more likely for them to be looking for residual "trace" from the atmospheric tests that doused the planet in "unnatural" residuals and odd isotopes.
The wikipedia article says "While animals have lived in the cave for 5.5 million years, not all of them arrived simultaneously. One of the most recent animals recorded is the cave's only species of snail, Heleobia dobrogica, which has inhabited the cave for slightly more than 2 million years."
For some reason, I feel sad. I was always fascinated when finding an article about newly found discoveries on Earth, and the comments from Hacker News were usually helpful and adding more to the stories.
Now, we talk about whether such discovery was written by LLM.
Pangram isn't useful for single documents. Too many false positives. It only gives something approaching actionable results if you use it on an entire corpus.
There’s residual and radioactive heating from the Earth itself that can provide energy for an ecosystem. We expect Jupiter’s moons to provide similar environments because of tidal deformations (which provide enough energy for Io’s vulcanism).
That has a lot more to do with whether the cave started out containing anything "more complex than insects." 5 million years isn't, like, an evolve-monkeys-from-scratch number of years. It's arguably an evolve-humans-from-chimps number of years. But if your cave doesn't contain any chimps... or even if it does but they all starve to death in Year 2 out of 5,000,000...
We went from wolves to chihuahua in just 40k years. Yeah selective breeding and all that, but it emphasizes that you can have very extreme changes in very brief periods of time. Imagine chihuahuas didn't exist today, and somebody said that within 40k years you could turn a wolf into one. It'd sound absurd!
Biology is quite arbitrary. For instance the dingo is basically just a wild dog, looks like a dog, can breed and create fertile offspring with dogs, yet is considered its own species.
So I prefer things that are more visceral in nature. There are lots of dogs that look a lot like wolves, but going from a wolf to a chihuahua is just kind of mind boggling in terms of the degree of change over such a small window of time. There's one extremely disappointed great great great ... grandfather wolf somewhere. 'Kids these days.'
Add to this that life in darkness sustained by microbes consuming a rock surface will be rather slow. Because any fast processes would just flash out after a boom/bust cycle or two. And presumably not just slow in terms of no cheetah vs antelope racing, but also much slower than out in the sun in generational turnover.
I wonder how big that cave system is, as in number of barley connected caverns. There might have been quite a few cycles one cavern flashing out and then slowly getting repopulated from the neighboring cavern(s). The "magical" balance we admire in nature isn't a miracle, it's the result of endless series of failure.
Yes, but an alternative would be that if an organism is perfectly fit for its static environment (that hasn't changed in 5 million years) then there is no benefit to evolving novel features.
More likely a dynamic environment is most important for intelligence.
> Oxygen levels run at roughly half of what our lungs expect. Carbon dioxide is present at concentrations enormously higher than in the outside atmosphere. The air is laced with hydrogen sulfide, the compound responsible for the smell of rotten eggs, and with methane and ammonia.
So it was not truly sealed. A truly sealed cave would have had no oxygen, it would have all been consumed in reactions with methane.
>Yeah it has significant groundwater inflow (like 5L/s).
where did it say this? i did a double-take because the article says the opposite, and i checked the wikipedia and didnt see it there either. is there another source of info? or am i being blind?
> Mesothermal, sulfidic groundwater (up to 1 mM in H2S) originating from an artesian sulfidic aquifer underlying the region enters at the bottom of the cave’s lower passages. This results in a constant flow occurring tens of centimeters to 1 m below the water surface at flow rates of about 5 L/s
This already happened in the past: they found a cave totally sealed since so long that spiders still had eyes, but non-functioning ones (for they were totally in the dark for generations and generations, eyes didn't make sense anymore).
Just as the holy books predicted. Oops, sorry, just as Darwin predicted.
Speleology aside, just a few days ago in one of HN discussions I was told LLM-produced content is already indistinguishable from human-made. But I'm sure I would have pretty good chances to correctly guess even the exact brand which helped to write this sloppy text. And it's the same with many stories I read.
Stake the claim then. I’ve also read a fair bit of LLM text and this didn’t trip anything for me.
Edit: after reading the author bio and the other articles, I agree with this assessment. Though I will admit, this particular article didn’t jump out as AI to me.
It's Claude. I picked up on it too (I am also a heavy Claude user)
The title alone gives it away. Sentences like this also trigger my daily annoyance:
"Researchers found an elegant way to test it, and the timing was almost too convenient."
Me: "Almost too convenient is expressing an opinion. Rewrite this to be straight forward, succinct information"
Claude: "You're right, and I was wrong. That's the second time you've had to correct me this session. Let me update my memory to adjust for this"
Eye twitch
And the nonsensical, chatty slang
> Let me ...
As if Claude actually needed the human user's permission or cooperation to do so.
I know what you mean. Flowery and fluffy. I don't see any clear giveaways, but it feels like it and is a bit annoying to read because the information in there isn't very dense.
the "why it matters" subtitle near the end seems very much like claude to me
That falls into the general category of 'authority groping' (yes, I coin my own terms) where a prediction engine without real reasoning, must apply numerous tricks to convey confidence, authority and substance where there may be less or none. Similar to the constant negative-propositions or "not" statement, where usually one to three things that do not apply are articulated, which has the result of credibility accumulation, I suppose, by showing "hey, I am right about not just one thing, but four!" hence the saturated styles we tend to see.
I saw a great video on YouTube by languagejones, and he asks: do we have to tell the difference between an LLM and an idiot?
The writing is just bad. I think this is clear to everyone, so I guess we could stop at that
"summary, a, b, c, why it matters" + fillers
like the summarization tools based on those models except it's the opposite, it was fluffed up into a long article
Just because the cave is sealed now, does not mean it has been sealed for millions of years. Even for geology, 5.5M years is a long time. I don't understand how they can claim it never had connections to the outside, just because it doesn't have connections now.
"Combined with the geology and the peculiar chemistry of the cave’s water and air, the evidence pointed to a system that had been closed off since roughly five and a half million years ago."
That quote, from the AI generated article, simply asserts the cave was sealed, it doesn't explain how the cave came to be sealed, nor how it is that modern humans infer the cave has been sealed and for how long.
On the flip side, the wiki article for the cave mentions that new species arrived as recently as 2 million years ago...
Geology is most probably largely how the sealed claim has been made .. it was open to the world when the limestone was forming, formed (Speleogenesis) by sky water (rain) permeating down and eating out a limestone void, and "sealed" by thick and impermeable layers of clays and loess (windborne sediment) some 2.5 million years ago during the Quaternary.
There are also genetic drift estimates, which put the "last" outside species to enter at about that time (
~ https://en.wikipedia.org/wiki/Movile_Cave )
Mentioning the recent Chernobyl accident is weird too. Surely an accident from long ago that is expected to be detectable is far more helpful.
All they proved is that nothing had entered since the Chernobyl accident.
Have there been many nuclear reactor meltdowns prior to Chernobyl?
Before 1986 there have been twelve "public" core meltdowns, and at least six, likely more, incidents not public on nuclear subs, etc.
~ https://en.wikipedia.org/wiki/Nuclear_and_radiation_accident...
You should look up how many live armed nuclear bombs have been accidentally dropped on mainland US soil.
More significantly, there's been a lot of above ground and atmospheric tests that have draped residuals across the planet, and a bunch of dirty Russian / former USSR sites that drift trace across Europe.
Isn't there a nuclear bomb sitting somewhere in a swamp in the USA that everybody lost?
"Yes"
- TBH I'd be surprised if it's still "lost" - not further reported on because [ XXXredactedXXX ] is more likely.
Also, if the finding is never recorded, it can be used to frame just about anyone who can be accused of finding it and smuggling it out of the US.
You could just look for isotopes from the atmospheric nuclear tests that were conducted? Doesn't need to be a reactor I would assume?
Indeed. That's one of the reasons why the "Before Present" time scale (that uses 1/1/1950 as the epoch) is sometimes called "Before Physics" (another one being that 1950 is less and less "present", of course)!
> that uses 1/1/1950 as the epoch
I guess we can say this is when humankind reached adulthood, became able to drink and got a driver’s license. And a gun.
Yes.
But also, finding something from within the last hundred years doesn’t say an awful lot about something claimed to have been a closed system for millions of years.
It's an AI article.
Looking for Chernobyl signature in cave during the late 1980s / early 1990s does seem daft - far more likely for them to be looking for residual "trace" from the atmospheric tests that doused the planet in "unnatural" residuals and odd isotopes.
The wikipedia article says "While animals have lived in the cave for 5.5 million years, not all of them arrived simultaneously. One of the most recent animals recorded is the cave's only species of snail, Heleobia dobrogica, which has inhabited the cave for slightly more than 2 million years."
> this was not an ordinary cave. It was something far rarer
so probably a cheap openai model
For some reason, I feel sad. I was always fascinated when finding an article about newly found discoveries on Earth, and the comments from Hacker News were usually helpful and adding more to the stories.
Now, we talk about whether such discovery was written by LLM.
Does it really matter?
Does anyone have the original source of this information? This article seems to be AI generated: https://www.pangram.com/history/fdf6dfd6-1a9c-4267-a90c-a3cf...
while looking at the wiki (https://en.wikipedia.org/wiki/Movile_Cave), this popped up from 2015 from bbc earth: https://web.archive.org/web/20210531110415/https://www.bbc.c...
Pangram isn’t a reliable detector
Still the linked article reads like slop
Doesn’t mean it is.
I bet you can find pre-AI articles that sound like slop too.
I keep requesting this here, but so far I haven't been given any links.
There's probably a good reason for that.
Pangram isn't useful for single documents. Too many false positives. It only gives something approaching actionable results if you use it on an entire corpus.
An updated Turing Test is overdue or has someone suggested one?
Isn't this the opening to a few of those "virus destroys humanity" movies?
Interesting that being sealed for 5 million years did not yield any life forms more complex than insects.
Maybe sun and space are importance factors for intelligence.
the sun is energy so i’d expect there to be a correlation there.
There’s residual and radioactive heating from the Earth itself that can provide energy for an ecosystem. We expect Jupiter’s moons to provide similar environments because of tidal deformations (which provide enough energy for Io’s vulcanism).
That has a lot more to do with whether the cave started out containing anything "more complex than insects." 5 million years isn't, like, an evolve-monkeys-from-scratch number of years. It's arguably an evolve-humans-from-chimps number of years. But if your cave doesn't contain any chimps... or even if it does but they all starve to death in Year 2 out of 5,000,000...
Humans didn’t evolve from chimps.
No but they could over 5 million years
We went from wolves to chihuahua in just 40k years. Yeah selective breeding and all that, but it emphasizes that you can have very extreme changes in very brief periods of time. Imagine chihuahuas didn't exist today, and somebody said that within 40k years you could turn a wolf into one. It'd sound absurd!
Isn't it more accurate to say we went from wolves to dogs? Dogs are not a group of species. The domesticated dog is a species.
Also canines seem to hybridize well, suggesting they aren't truly differentiated anyways.
Biology is quite arbitrary. For instance the dingo is basically just a wild dog, looks like a dog, can breed and create fertile offspring with dogs, yet is considered its own species.
So I prefer things that are more visceral in nature. There are lots of dogs that look a lot like wolves, but going from a wolf to a chihuahua is just kind of mind boggling in terms of the degree of change over such a small window of time. There's one extremely disappointed great great great ... grandfather wolf somewhere. 'Kids these days.'
Same thing happens with plants on a similar timescale right? Cultivars look nothing like their wild relatives at this point.
Add to this that life in darkness sustained by microbes consuming a rock surface will be rather slow. Because any fast processes would just flash out after a boom/bust cycle or two. And presumably not just slow in terms of no cheetah vs antelope racing, but also much slower than out in the sun in generational turnover.
I wonder how big that cave system is, as in number of barley connected caverns. There might have been quite a few cycles one cavern flashing out and then slowly getting repopulated from the neighboring cavern(s). The "magical" balance we admire in nature isn't a miracle, it's the result of endless series of failure.
Yes, but an alternative would be that if an organism is perfectly fit for its static environment (that hasn't changed in 5 million years) then there is no benefit to evolving novel features.
More likely a dynamic environment is most important for intelligence.
Great first line for a story.
It is indeed well written but I feel it has LLM smell, not that is a bad thing.
Let’s not be racist, please. It doesn’t matter if the intelligence is based on carbon or silicon.
You think machines are a race? Where did you go to school?
Even space is a race.
> Oxygen levels run at roughly half of what our lungs expect. Carbon dioxide is present at concentrations enormously higher than in the outside atmosphere. The air is laced with hydrogen sulfide, the compound responsible for the smell of rotten eggs, and with methane and ammonia.
So it was not truly sealed. A truly sealed cave would have had no oxygen, it would have all been consumed in reactions with methane.
Yeah it has significant groundwater inflow (like 5L/s). Nothing about this is sealed.
>Yeah it has significant groundwater inflow (like 5L/s).
where did it say this? i did a double-take because the article says the opposite, and i checked the wikipedia and didnt see it there either. is there another source of info? or am i being blind?
> Mesothermal, sulfidic groundwater (up to 1 mM in H2S) originating from an artesian sulfidic aquifer underlying the region enters at the bottom of the cave’s lower passages. This results in a constant flow occurring tens of centimeters to 1 m below the water surface at flow rates of about 5 L/s
https://pmc.ncbi.nlm.nih.gov/articles/PMC10672346/
Right, so it was sealed only in the biological sense, which makes sense since there needs to be an inflow of energy
It's all fun and games until they run into another branch of homo sapiens ala the descent....
Or the Reptilians. Or nazis from the centre of the Earth.
> Workers Drilling in Romania Broke into a Cave
Oh, so that's why my neighbour, Vasile, drills day and night. He hopes to discover a new life form. /s
This already happened in the past: they found a cave totally sealed since so long that spiders still had eyes, but non-functioning ones (for they were totally in the dark for generations and generations, eyes didn't make sense anymore).
Just as the holy books predicted. Oops, sorry, just as Darwin predicted.