This paper shows exactly how good “Peer Review” is
It’s not just that a clever AI image slipped through peer review, it’s that it was garishly fake in a supersize kind of way. Scientifically everything about it was radioactive satire and yet it still got through “peer review”. The words are gibberish. The editors didn’t even run a spell checker on it before publishing it, let alone the gaze of a single trained biologist in the field.
The paper has been retracted thanks to the real peer review which happened on social media. This was a case of X (formerly Twitter) saves the day. Where normal peer review can take up to two years (if you are an unpopular skeptic) it was only three days from the X review to retraction.
The Telegraph sums it up:
A scientific paper purporting to show the signalling pathway of sperm stem cells has met with widespread ridicule after it depicted a rodent with an anatomically eye-watering appendage and four giant testicles.
The creature, labelled “rat”, was also sitting upright in the manner of a squirrel, while the graphic was littered with nonsensical words such as “dissilced”, “testtomcels” and “senctolic”.
FIGURE 1
Spermatogonial stem cells, isolated, purified and cultured from rat testes.
The macrophages have become Macromages. The Natural killer T-Cells have become “nokillas”, but it’s not a name-swap, it’s just complete and utter nonsense — like a microbiology word soup met a UFO.
The JAK/STAT pathway and immune regulation in spermatogonial stem cells
The researchers even told them the images were faked: “(Images in this article were generated by Midjourney).”
At best, perhaps this is a real paper with junk AI images. There are no obviously imaginary words in the text (unlike the graphics). But it’s still a devastating take on “peer review”. I mean, these images are practically satire… If rushed scientists are using AI to help them write, and AI to get cheap images, and the Peer review journals are just posting anything without even checking, modern science is a zombie.
This is not a one off problem, and the use of AI to create images in peer reviewed science is widespread:
[The Telegraph] Writing on the Science Integrity Digest, Dr Elisabeth Bik, the Dutch microbiologist who works spotting manipulation in scientific papers, said: “Of course, we can have a good laugh at these figures, and wonder how on earth the handling editor and the two peer reviewers didn’t catch this.
“These figures are clearly not scientifically correct, but if such botched illustrations can pass peer review so easily, more realistic-looking AI-generated figures have likely already infiltrated the scientific literature.”
Dr Bik has identified more than 1,000 papers which have fraudulent imagery, most of which she believes was generated by AI.
Amanda Yeo claims as many six real people supposedly gave it a tick:
It isn’t clear exactly how these diagrams made it all the way to publication without being picked up. The article was edited by a member of Frontiers‘ editorial team as well as reviewed by two other parties, which means at least six people gave it their approval. In a statement to Motherboard, one of the reviewers said that he had only assessed the paper for its scientific aspects, and that it was not his responsibility to check the accuracy of the AI-generated images.
For what it’s worth, Frontiers in Cell and Development Biology (Front Cell Dev Biol) is not a big name in the medical world, but they have apparently published 10,730 papers in the National Library of Medicine. The umbrella publishing unit called Frontiers –says it is the 3rd most cited publisher.
The retraction message isn’t exactly confidence boosting:
Following publication, concerns were raised regarding the nature of its AI-generated figures. The article does not meet the standards of editorial and scientific rigor for Frontiers in Cell and Development Biology; therefore, the article has been retracted.
This retraction was approved by the Chief Executive Editor of Frontiers. Frontiers would like to thank the concerned readers who contacted us regarding the published article.
The lead author Xinyu Guo supposedly works at the Department of Spine Surgery, Hong Hui Hospital, Xi’an Jiaotong University, Xi’an, China, and has published some genetic sort of research. Maybe they are real? Hard to say without digging deeper.
REFERENCE
Citation: Guo X, Dong L and Hao D (2024) Cellular functions of spermatogonial stem cells in relation to JAK/STAT signaling pathway. Front. Cell Dev. Biol. 11:1339390. doi: 10.3389/fcell.2023.1339390
*Title changed in an effort to encourage more on topic comments.
Ties in well with a piece from The Conversation.
200
But I saw it on TV…
So it must be true.
130
I have decided!
There is no such thing as Artificial Intelligence. No matter how wonderful a computer may be, it can only tell us what we have told it.
The change that we are seeing has been brought on by a dramatic reduction in the cost ofcomputer hardware, which has been happening ever since the 1960s when Gordon Moore made his famous prediction.
It is spine tingling to see that Moore’s Law may finally expire within a couple of years after Gordon himself.
90
No such thing as intelligence. Human can only tell us what a human has been told. These reviewers are a case in point.
22
My definition of intelligence is the ability to relate cause and effect.
10
Amazing but not surprising.
Back in the day, scientific papers were genuinely peer reviewed but I think peer reviewers have become lazy, incompetent and terrified to express an alternative opinion (on anything) since some time in the 1990’s.
It also coincides with the heavy promotion of the lie by Klimate Kultists that scientific fact is decided by “consensus” and also that there is such a thing as “settled science” (sic).
We again saw the lie of “science by consensus” and “settled science” repeated with the covid vaccine disaater.
So-called “peer reviewers” represent “the consensus” and are part of the establishment. If they had alternative ideas beyond “settled science” and consensus in general, they wouldn’t be invited to do peer review. E.g. obviously no independent thinker who would question the anthropogenic global warming fraud or the “settled science” that “covid vaccines are “fully safe and effective” would be asked for their opinion – on any topic.
Even though peer reviewers don’t notice or ignore blatantly obvious errors and complete nonsense, I’d be willing to bet they would reject any paper that questions the Official Narrative in regard to supposed anthropogenic global warming or the safety and efficacy of covid “vaccines”. In fact, we know such papers have been rejected and scientists who attempt to publish such opinions get canceled, censored, sacked/fired, defamed and libeled by their “peers”.
361
Humorism – This is no joke
“Science by consensus” can be entrenced for a long time. The “consensus” that “humors” or bad smelling air was the cause of illness lasted for around 2000 years (Humorism).
It required real science based on evidence gathering and interpretation to finally defeat this “consensus”.
An example being John Snow who is considered one of the founders of modern epidemiology and early germ theory, in part because of his work in tracing the source of a cholera outbreak in London’s Soho, which he identified as a particular public water pump.
Today we have a new “consensus” that something in the air causes problems for mankind and this consensus is supported by computor modelling without the application of proper scientific inquiry.
“The Science is settled” and anyone who dares to question the “consensus” is labelled a “denier”.
260
“Because of his work in tracing the source of a cholera outbreak in London’s Soho, which he identified as a particular public water pump.”
By removing the handle from the pump, cholera deaths dropped dramatically.
220
Four years latter The Great Stink occurred in an event in Central London during July and August 1858 in which the hot weather exacerbated the smell of untreated human waste and industrial effluent that was present on the banks of the River Thames.
By June the stench from the river had become so bad that business in Parliament was affected, and the curtains on the river side of the building were soaked in lime chloride to overcome the smell.
This prompted the politicians to fund a major upgrade to London’s sewerage system lead by Joseph Bazalgette which included embankments along the Thames which are a feature of London visible to tourists
The Crossness Pumping Station has been descibed as “a masterpiece of engineering – a Victorian cathedral of ironwork”.
https://crossness.org.uk/visit/
240
And the butchery of the prefrontal lobotomy was also once “consensus science”. One of its pioneers, António Egas Moniz, even won a Nobel Prize for it.
181
I only follow beer reviewed science.
380
Honk,
We have the “Pub Test” which is generally accepted . It generally means that something has to make sense . Commonsense is no longer common .
180
It could be that the UNIPCCC was involved in the cancellation of the paper because it contradicts their position on the urgent need to strive for lower emissions.
Clearly this rat would not conform to current policy.
200
Here is an old story demonstrating that the world of peer reviewed publishing has been broken for a long time.
Some readers may remember the time when some MIT students wrote a “nonsense scientific paper generator” and the nonsense so produced passed peer review…
(REST IS PAYWALLED)
211
Scientists have made themselves a laughing stock and unreliable. Some by publishing rubbish in support of the climate change scam but most by failing to call out the former. Truth eventually wins out simply because nature intervenes. The news this morning is about the failure of the BoM to accurately predict the weather for the current summer. They were desperately telling us that Western Australia had its hottest day ever a few days ago while trying just as desperately to hide the fact they had predicted a super El Nino with high temperatures and low rainfall which caused a sell off of sheep and cattle. That has now been followed with restocking to use the extra grass. Small beans to the bureaucrats but a disaster for pastoralists many of whom now subscribe to independent weather forecasters. When weather is the prime driver of profit or failure the BoM can not be trusted to give good advice.
370
Now they’re trying to blame it on TV media and social media for ‘distorting the message’. LOL
200
One of the most infamous nonsense paper hoaxes was by Alan Sokal, a physics professor at New York University and University College London, in 1996.
230
So the real peer review is by the consumer? Interesting.
110
And this combination of AI and “peer review” is widespread. I added a few quotes to the post belatedly from Elizabeth Bik who found a thousand cases of fraudulent images created by AI.
10
Bwahaha! The Science™ sucks.
Humour saves humanity yet again.
130
I thought Mann’s HS and their cut and paste “extra warming” after 1960 was about as stupid as we should expect, but that Rat was something else.
But good to see that at least some of their BS and laughable lunacy are quickly exposed by the common sense of the general audience and their comments at X.
180
It’s interesting that the corruption of the scientific method and the peer review process by the Left primarily benefit two major, highly profitable (for the Elites) hoaxes of our time related to supposed anthropogenic global warming and the supposed efficacy and safety of covid “vaccines”.
Not only are these hoaxes highly profitable for some, they enable the imposition of unprecedented Government controls and removal of human rights from we, the people.
190
These revelations all lead to a loss of confidence and respect for science which can never be good for society.
The loss of confidence in science only benefits the purveyors of fairy tales of the Left for their own personal benefit of power, control and money via their lies related to frauds such as supposed anthropogenic global warming and covid vaccines.
140
It’s only when the loss of public confidence is complete, that real scientists will regain authority over their field and return to real science. Up until then it’s just “turtles all the way down”.
50
Secretive and Subjective, Peer Review Proves Resistant to Study
JENNIFER COUZIN-FRANKEL
SCIENCE 20 Sep 2013 Vol 341, Issue 6152
p. 1331
DOI: 10.1126/science.341.6152.1331
https://www.science.org/doi/10.1126/science.341.6152.1331
110
Sounds like a publication from Monty Python! Biggus Rattus.
150
It is not a rat Mr. Fawlty. It is hamster with a biguss dickuss.
40
I can just see that in my mind’s eye!
20
What was the peer review process and did it include the illustrations?
010
Perhaps the reviewers were AI too…
100
No doubt that is actually possible in some sense.
This is a junk journal and I think it is hyperbole and lazy to use this as an example of the problems with peer review. Peer review is very problematic and is discussed a lot by the science community and change is happening though too slowly and without consensus (use a search engine for “history of peer review”).
A separate issue is the emergence of junk journals and citation pumping. These journals spew out crap that gets cited by other crap publications for benefits to who or what I can’t grasp but certainly to the detriment of science. This is a more bizarre and unsubtle example. I don’t know what the solution to this is or that any action is being taken that will address it.
27
Gee A.,
“for benefits to who or what I can’t grasp ”
Many academic types have to have an “end” degree (like a Ph. D.) and also publish in journals and, perhaps, bring in grant money (supports graduate students and the school).
A request of funding will include a certain percentage that supports the school’s costs for heat, light, repairs, administrators, and unknowns. One place I know of wanted as many as 6 publications each year (or a reason why not) for the person to get raises and advances through the ranks.
80
Perhaps for a small number of cases in some jusistiction that has poor standards. in general, what you suggest wouldn’t work with something published in that journal s it is essentially unranked. Also, the institution they currently work for would not permit their name to be associated. Funding bodies rank journals and Ph.D reviewers are tough reviewers. I think that there is not enough money in it from those sources for someone to create such a journal – except it might cost them essentially nothing and take no effort to maintain so why not try to hook in some suckers.
04
Hyperbole? The “junk” journal has 10,000 papers in PubMed. And Elizabeth Bik found a thousand cases of fraudulent images created by AI in other journals.
Do you admit AI and peer review have a problem, or want to keep pretending it’s just this one journal that got unlucky.?
50
Do I admit? Yes. No need to ask this if you read my other comment that specifically states this.
01
Good question.
The main problem with the paper appears to be diabolical OCR for the text on the diagrams. That would not be the result of AI-based Midjourney but more likely image compression by the publisher. Perhaps the reviewers saw the original diagrams.
10
A lack of oversight in late production makes sens but it amounts to the same in the end. A poor process results in a laughable outcome.
12
Leo, I don’t think the nonsense words are just an image compression thing. The actual art, the arrows, the flow, the structures do not seem to make sense, even if they were labeled correctly. Perhaps I’m missing something…
110
it is total nonsense
00
When the paper’s second author is named Dong L, perhaps you should become suspicious.
Or was that the name of the rat?
160
I knew of the name Long Dong but one bloke reportedly had the initials E.R. and that caused a lot of ribbing.
40
THat’s unfair. He was probably sensitive about it.
33
“Donger” was very sensitive about it.
40
Repeating a joke and making it out to be your own is not cricket.
03
I don’t play or watch cricket.
10
Yes but what about the other side that is described by some people rudely?
41
I put a lot of blame on the redefinition of a scientist. I remember well Prof Flim Flannery holding forth on hot rocks (the technology is straightforward), Nuclear power (at times he took both sides) and Climate (“even the rains which fall will not fill the dams”). Being completely wrong never fazed him. Like the BOM.
And the corruption and marketing of PhDs. I sat through an RMIT ceremony where they awarded 200 PhDs in one night. One PhD was on the logistics of truck distribution of Halal meat in Australia. People now say they would have done PhD but didn’t have the cash or the time. It used to be a very exclusive club of high performance scientists where you had to be invited. Now it’s a commercial decision or life choice?
Even computing now has Computer Science as a PhD and with no necessary training or expertise in chemistry, mathematics, physics or any hard traditional demanding science, they are ‘computer scientists’. I have run into these people who hold forth on subjects they have never studied or in which they have any proven skill. Being able to write software does not make you a scientist.
Our former Chief Scientist is utterly convinced of the truth of man made Global Warming, despite being an electronic engineer. And his company writes science curricula for schools.
And with the miracle of Dr. Google and AI, people can create or find papers like this out of thin air to refute real scientists. And these ratbag papers even get published, usually as a rebuttal to real science which dares question fake science. There is a raft of rebuttals now to Prof Will Happer’s plain science proving that CO2 is saturated as a Greenhouse gas. It’s hard to find the real scientists among the fakes.
AI means you can produce a supporting paper on any subject on request. And as Jordan Petersen found, it will even fake very credible references which don’t exist! AI will casually fabricate lies if necessary.
What then is the future of real science and real scientists? Fake scientists are bad enough. Fake papers destroy any belief in peer review, which after all is not about the truth but about the method. And is often a rubber stamp anyway.
So there are now tens of thousands of scientists with PhDs who are experts on everything, like Flannery. And Tim could not get into any science course, so he took an English degree at La Trobe in 1974, at a time when it was a university of last resort in the days before every technical college became a university.
Now we are told what to think by arbitrary self appointed experts who call themselves scientists. Australia’s Climate Council had only one real scientist, the late Physical Chemist Prof Will Steffen who effectively told us that we would have to cover half of Victoria in solar panels to power Melbourne, during the day time. He was right.
Real scientists keep to themselves while publicity hungry public scientists like Michael Mann refuse to even release the original tree ring data which cost the world trillions. He says he owns the data, despite the costs being born by the state. I met a man in the street who proudly called himself a ‘Climate Scientist’. As far as I know his only qualification was an interest in reading articles about the climate. This redefines AI.
Perhaps the core question is why the United Nations has any business in the weather in the first place. And why every government department since has pushed climate change as the most important science in the world. Even President Joe Biden says it is a greater threat than nuclear war. Except none of it is true.
380
Well said TdeF.
And the devaluation of the PhD is truly terrible.
I can think of no tertiary institution in Australia or the West that actually practices genuine scholarship any more.
I have been looking for one (or more) so I can bequeath them something in my will but haven’t found one yet.
160
Try the JC II Institute.😉
80
Isn’t JC II the Second Coming? (If that isn’t totally inappropriate given the original theme of this thread, as if I give a rat’s)
70
I’ll do the funnies here, although they’re underRATed. 😁
60
Sorry, didn’t mean to rattle you
40
This decline in the validity of PhDs is well illustrated in this article, as one can now do a PhD on Taylor Swift fans. The article shows how this pop star’s songs are listed on some university canons alongside Aristotle and Shakespeare. If you require any further proof the intellectual standards of tertiary academia are going down the toilet, click on the link below
https://www.smh.com.au/national/pop-star-philosopher-poet-taylor-swift-is-shaking-up-how-we-think-20240207-p5f342.html
80
A decline indeed – AC/CS’s lyrics are more insightful for our times
60
Taylor who?
90
Wow! All this from a junk journal article. I think you might be over extrapolating.
A wee bit.
36
Extra pole ? 😆
50
To highlight the oversights in one peer reviewed “junk journal” doesn’t equate to stating all peer reviewed articles contain errors and flawed information. But it certainly does bring into question any published article or study that is declared as peer reviewed, a term synonymous with words such as ‘authentic’ or ‘verified’.
During the pandemic many studies involving repurposed drugs to treat Covid which showed promising results were routinely dismissed because they were not “peer reviewed”. Case closed- the study doesn’t have the required gold label attached to it, sweep it under the rug.
So if you’re going to imply one seemingly ridiculous junk journal article doesn’t smear all other peer reviewed articles, please also entertain the idea that many articles or studies that are not peer reviewed may well contain valid content worth examining.
Apologies to all other readers for bringing up Taylor Swift in the thread, my bad 😉
30
“Computer Science” is quite an old term, which covers quite a few sub-disciplines.
To be fair, most of us end up in something more akin to Engineering (but with more of a propensity towards tooled leather boots, spurs and big hats), but there is a strong theoretical applied mathematics basis. You just have to read Donald Knuth’s works to see that.
Similarly, “Veterinary Science” is an even older term for a specialised area of the life sciences (an older term still).
40
A Ph.D has never been a science degree.
07
You might want to tell Sydney Uni that 🙂
40
The Title over old Physics at Melbourne University remains not physics but ‘Natural Philosophy’, what we also call Rational Science. It was the philosophy of Rene Descartes, cogito ergo sum. And it created the modern world.
Prior to Descartes science had a different sense of just knowledge and not necessarily understanding. Like medicine which was awash with quacks. As for Climate Science, these people are not even meteorologists, a strict science discipline which is quite old and currently ignored. Climate Science is more Climate religion, also populated by quacks.
50
We need a new version of Bovine excrement bingo . This study (and current events) take it to a new level….
90
It’s just a reminder to people to keep your Viagra pills away from food sources. 😆😆
90
At the bottom of the disc, presumably a testicular cross section from a well endowed rat, is wonderful nonsense, a parody of science nomenclature
IMMOUMINOMUDUODIUILATIUCIATION -> Ingrim mrology signarg polocllay.
It says it all. Anyone who thought this paper was real did not read it or even look at the pictures.
100
I take it back. The funniest label is RAT ————>
60
“The case against science is straightforward: much of the scientific literature, perhaps half, may simply be untrue. …. science has taken a turn towards darkness. As one participant put it, “poor methods get results”
Richard Horton @TheLancet editor. 2015
110
As an old programmer (very old) I fear that the hype around these early LLM AI models is such that people have forgotten the premise of GIGO or Garbage In = Garbage Out.
120
Punched cards (I can still remember the smell of the cards and the oil used for the machines)…
80
You win! 😆
30
Odour memories!
50
Did you have an IBM 1620 and use FORTRAN?
Friden desktop calculators?
50
Eliot 803, using Autocode and machine code and a Monroe Calculator and another calculator whose name I forget (began with an ‘E’)
20
IBM 1130 and FORTRAN 16K words (32K bytes) memory
1 megabyte “hard drive”
initially printouts on a IBM 407
soon replaced by a 1132 printer (release date 1965)
Redefine “old”?
20
Control Data Cyber 6600. It was a supercomputer with 130,000 words of memory (each 60 bits long).
10
Geez, I’m starting to feel privileged, in the mid 80s we had Apple IIe computers at school for lessons. The green neon characters on the black background are reminiscent of The Matrix code displayed in the movies. Fitting, l feel like l took the red pill only a couple of years ago, much more aware of what’s going on in the world now for better or worse
20
Perforated tape in my day.
60
You youngsters. What’s wrong with Mercury Delay Lines?
70
Lucky! We only had unpunched tape. And had to buy bags of holes.
50
Buy bag of holes?
Luxury!
We had to go into woods and pick our own.
20
The first thing we learnt was GIGO!
20
None of these fraudulent climate “models” would exist today if GIGO was still taught, plus the idea that any model had to be first validated with real world data before use.
60
Frontiers Media already have a reputation as a publisher of predatory journals, which take advantage of authors by asking them to publish for a fee.
https://predatoryreports.org/news/f/list-of-all-frontiers-media-predatory-journals
It’s a multinational, founded in Switzerland. Their journals are open access so are funded at least partly by payments from authors; thus there’s incentive for the editors to publish as much as possible, never mind the quality. When the authors are from a Communist dictatorship, that risk is increased.
60
I guess it was published following peer review by some other rats.
60
The paper will be probably resubmitted with a disclaimer that the rat was actually born female but underwent mandatory trans surgery to comply with university requirements 😉
70
Looking at it again, are we quite sure this wasn’t intended to be an early April Fool’s joke?
30
No.
10
Hmmm……….rats with big dicks; sounds like an allegory for the political class.
Jonathan Swift & Thackeray would be unimpressed but it is appropriate for pollies on offer today.
50
What was the world like before Scientific Revolution?
Limited scientific knowledge: Before the Scientific Revolution, there was limited understanding of the natural world and how it worked. People relied on religion and superstition to explain natural phenomena, and there was little scientific inquiry or experimentation.
It seems that the woke Western world has gone back to “People relied on religion and superstition ( Gretas Gaia Climate change alarmists) to explain natural phenomena, and there was little scientific inquiry ( Timmy Flannery no more rain) or experimentation.” While the woke Western world slowly destroys itself, the rest of the world is going hard with fossil fuels.
70
Here is one of the original “spoof” reports, mainly engineering.
It is as funny today as then, late 1970s.
Geoff S
https://www.youtube.com/watch?v=Ac7G7xOG2Ag
50
We had one in the garage. Amazing technology but it didn’t work.
30
That is one happy rat. Or at least he was before bits got disected.
That other thing is a Pizza Hut try-out for those for whom the Hawaiian is not quite outrageous enough.
10
Hey Siri … print me a PhD wouldja?
30
[…] By Jo Nova […]
00
There is no such thing as AI.
The original and competing search engines in the early years of a burgeoning Internet were called artificially intelligent because of how they developed methods enabling a user to creep ever closer to the perfect search entry for identifying what they were searching for. A simple example would be Fred Pope, non-religious which could ferret out all the Vatican based potential candidates.
There seriously hasn’t been a time when AI wasn’t present in programming syntax and algorithms e.g. repeat this ten times while x is less than 1500 makes several outcomes possible with different pathways from thereon.
The AI problem as I see it is the very long in the tooth desire to impress where hyperbole is the means to an end. It’s as if the tech companies need a reason to keep computers in fashion least smartphones reduce expectations and old software (which are still the best programs you will find anywhere) provide many examples of exemplary coding of pathways from A to Z visiting some or all points in between if required. There are many classic examples in the deep, dark web.
As the guy said on my first day with computers – they are completely dumb unless you know what you are doing – and you can check this out by misspelling your name to start with. So no intelligence at all in a machine that, like a smartphone, needs a lot of confidence to overcome fears and anxieties but the more you do the dumber the actual machine really becomes in your head.
60