Use of Artificial Intelligence in Research

Guest Post by Willis Eschenbach

As an amateur scientist, I am all in favor of one aspect of the emergence of artificial intelligence (AI). I employ it as the world’s fastest librarian, able to gather and summarize an amazing amount of data in a few moments.

Recently I thought I might use AI to educate me on the science of autism and vaccines. Note that I came into this question knowing almost nothing about any underlying scientific studies of the possible connection between autism and vaccines, other than numerous passionate claims from both sides of the question on the web.

Here’s how I went about gathering and checking the information. Thanks to a heads-up from Charles the Moderator, I use an AI called perplexity.ai. I’ve stuck with it for a curious reason—perplexity once told me it was reluctant to answer a question I asked about rainfall in a remote part of the Sahara desert because there was so little data available that it might be hallucinating an answer … I took that as a good sign that it recognizes that it has limitations.

My questions are in bold below, my comments are in plain text, and perplexity’s answers are in italics.

Me: List major studies showing that autism either is or is not associated with vaccination.

Note the construction. I’ve found that in doing scientific research it is important to specifically ask for evidence for both sides of the question. You want to avoid general questions like “What is known about autism and vaccines”. With that kind of phrasing, often an AI will not fully explore the question space, leaving out important evidence for one side or the other.

Here’s perplexity’s answer to my question above:

Several major studies have investigated the potential link between vaccines and autism. Here’s a list of key studies showing no association between vaccines and autism:

  1. Taylor et al. (1999):
    • Examined 498 children with autism in North Thames, England
    • Found no difference in autism rates between vaccinated and unvaccinated children
    • No clustering of autism diagnoses after MMR introduction
  2. Madsen et al. (2002):
    • Studied 537,303 children in Denmark
    • No increased risk of autism in MMR-vaccinated children
    • No association between age at vaccination, time since vaccination, or date of vaccination and autism development
  3. Jain et al. (2015):
    • Evaluated about 100,000 children with older siblings with or without autism
    • No increased risk of autism spectrum disorder (ASD) associated with MMR vaccination
  4. Danish study (1991-1998):
    • Included over 500,000 children
    • No link found between MMR vaccine and autism or autism spectrum disorders
  5. Meta-analysis by Taylor et al. (2014):
    • Analyzed 10 studies involving over 1.2 million children
    • Reaffirmed no association between vaccines and autism
    • Suggested MMR vaccine may actually decrease autism risk

Hmmm, sez I …

Now, as I mentioned above, AIs do hallucinate. So my next directive to perplexity is:

Verify that each of those is a real study, not a hallucination.

I‘ll verify the studies mentioned in the search results:

  1. Brent Taylor et al. (1999): This is a real study examining 498 children with autism in North Thames, England.
  2. Madsen et al. (2002): This is a genuine study published in the New England Journal of Medicine, involving 537,303 children in Denmark.
  3. Taylor et al. (1999) reanalysis: This appears to be a real reanalysis of the original Taylor study, though I don’t have enough information to confirm its exact publication details.
  4. Hviid et al. (2019): This is a real nationwide cohort study in Denmark, published in the Annals of Internal Medicine, involving 657,461 children.
  5. Time-trend analysis using the UK General Practice Research Database: This is a real study, though specific authors aren’t mentioned in the provided information.

Noticing that one study was left out, I asked:

Verify that 3. Jain et al. (2015) is a real study, not a hallucination.

Yes, Jain et al. (2015) is a real study, not a hallucination. This study is mentioned multiple times in the search results:

  • It’s listed in the Children’s Hospital of Philadelphia vaccine education center information
  • The full citation is provided: “Jain A, Marshall J, Buikema A, et al. Autism occurrence by MMR vaccine status among US children with older siblings with and without autism. JAMA 2015;313(15):1534-1540.”
  • The JAMA (Journal of the American Medical Association) article details are given, including the publication date of April 21, 2015
  • The study’s methodology and findings are summarized, noting it evaluated about 100,000 younger siblings and found no association between MMR vaccination and increased risk of autism spectrum disorder
  • It’s referenced in a Science Daily article about MMR vaccine and autism

The study is also cited in an Immunize.org document listing evidence that MMR vaccine does not cause autism

These multiple, consistent references across different sources confirm that Jain et al. (2015) is indeed a real study examining the relationship between MMR vaccination and autism risk.

All of these studies appear to be real based on the information provided. However, for some studies, I don’t have enough details to fully verify their publication information or authorship.

Upon request, Perplexity also provided details for all of the other studies it found in its research, viz:

  • California study comparing MMR vaccination rates with autism diagnoses (1980-1994): This appears to be a real study, though I don’t have enough information to confirm specific authors or publication details.
  • Quebec study of 27,749 children from 55 schools: This is likely a real study, but I don’t have enough information to confirm specific authors or publication details.
  • Danish national registry study of 537,303 children (1991-1998): This is likely referring to the Madsen et al. (2002) study mentioned earlier.
  • Atlanta study comparing 624 autistic children with 1824 matched controls: This appears to be a real study, though I don’t have enough information to confirm specific authors or publication details.
  • English study of 12,810 children born 1991-1992: This is likely a real study, but I don’t have enough information to confirm specific authors or publication details.
  • UK study of 100,572 children born 1988-1997 using the General Practice Research Database: This appears to be a real study, though I don’t have enough information to confirm specific authors or publication details

I found it fascinating to see its judgment of the validity of the information it is examining.

Moving on, I asked:

Provide a list of key studies claiming some association between vaccines and autism

Based on the search results provided, there are no credible studies claiming an association between vaccines and autism. The search results consistently report that extensive research has found no link between vaccines and autism. However, one flawed study is mentioned that initially sparked the controversy:

Wakefield et al. (1998):

  • Published in The Lancet (later retracted)
  • Suggested a potential link between MMR vaccine and autism
  • The research was later found to be fraudulent
  • The lead author lost his medical license
  • The medical journal retracted the paper

It’s important to note that this study has been thoroughly discredited and numerous subsequent studies have found no link between vaccines and autism.

————-

Look, friends, I know that there are real reasons to be cautious about vaccines. These include the steep increase in the number of recommended childhood vaccines as well as the increase in the number of adjuvants, preservatives, and other components in modern vaccines.

However, from this examination of solid scientific evidence about the question, it is clear that autism is not among the reasons to be cautious. Well-designed studies have repeatedly failed to find any link between vaccines and autism, and no studies have demonstrated such a link. Pick another objection, that one’s done and dusted.

Finally, some folks say that vaccinations don’t really protect us against diseases. While this may well be true about the COVID-19 vaccine, here are the observed rates of cases and deaths before and after the introduction of vaccines for a variety of illnesses.

Zooming in close reveals interesting pre- and post-vaccine morbidity and mortality statistics. Cause of the most deaths before vaccines? Whooping cough aka pertussis aka croup, 30 deaths per million. Followed by flu/pneumococcal disease (24/million), polio (22/m), diphtheria (14/m), and tetanus (3/m).

And here is what Jonas Salk did with his polio vaccine.


So that’s the end of the evidence and the beginning of the discussion. First, I know that folks are passionate about this question. So please, don’t shoot the messenger.

What I’ve presented above is how to use perplexity or any AI to unearth scientific evidence about very specific questions of a) autism and vaccines and b) the overall efficiency of each vaccine. I take no further position other than advising caution on everything.

Next, the myriad of questions about vaccines will not be solved on this web page. So let me ask folks to not take sides in the comments. It won’t help. As I said above, other than those two questions of autism and efficiencies, other reasons for caution remain. Exactly what is in each vaccine? What concentration? When, where, and how was it tested?

So in lieu of taking sides, let me invite people instead to use AI to dig out answers to those particular questions that remain in your mind about some particular vaccine, to search for solid scientific evidence, and to set aside the impassioned web-based claims and position-taking in favor of an actual search for current scientific knowledge.

Here’s how to do it. Easy money.

Finally, one thing to emphasize. As the first graphic above shows, there is no one vaccine question just as there is no one vaccine. There’s no overall good/bad, black/white issue.

Instead, every vaccine and every component of every vaccine has its own story, its own effects, and its own unanswered questions. For example, in the US, in what may be an excess of caution, childhood vaccines contain no mercury, while the adult versions of the vaccines contain millionths of a gram. Different issues, different relevant questions. Another example. I lived through the time of polio and Jonas Salk’s miraculous vaccine. And I even worked for several months as the night attendant of an adult polio victim spending the rest of his entire life in an iron lung, paralyzed from the neck down … he was married with one child and had just finished his medical residency when he was struck down. I would read to him, and play chess with him … dear friends, you don’t ever want to see that happen to anyone, much less your child.

The Salk vaccine brought a huge benefit. I can guarantee we don’t want to go back to those days. Different vaccines, different benefits, different questions, different issues, there’s no one overarching “Are vaccines good or bad?” question.

My best regards to each of you, stay well out there in this lovely but deadly world,

w.

PS—I’m still persona non-grata on X, with my account improperly suspended as I explain in my blog post below:

I ask anyone on X to use any leverage you have to get the X censorship gods to unsuspend my account, @WEschenbach. If you could do me the honor of linking to my blog post above, it will allow my voice to be heard.

As Usual: I ask when you comment that you quote the exact words you are discussing, so we can all be clear about your subject.

Get notified when a new post is published.
Subscribe today!
4.7 32 votes
Article Rating
117 Comments
Inline Feedbacks
View all comments
January 4, 2025 10:34 am

Very interesting, Willis. Thank you. Now I will have to try Perplexity.

Reply to  Willis Eschenbach
January 5, 2025 4:57 am

Done! See farther below. My comment was held for moderation for quite a while.

Rud Istvan
January 4, 2025 10:37 am

Wakefield, a gastroenterologist, did enormous damage with his fraudulent MMR/autism study. It was based on examining just 8 children, all of whose parents were clients of the lawyer who engaged Wakefield while suing the MMR producer for many millions. Wakefield himself planned to make $30 million/yr by selling a ‘GI autism diagnostic test’—despite the fact that there is no even complicated possible connection between the gut and brain. Deconstructed as a major example in ebook The Arts of Truth.

Yet the fraudulent myth lives on. The reason is a correlation without causation. The MMR vaccine is usually administered between 12 and 18 months of age, and the earliest autism signs emerge is between 18 and 24 months.

As WE’s post vaccine results figure suggests, the COVID19 ‘vaccine’ was nothing of the sort. CDC redefined ‘vaccine’ to fit Pfizer and Moderna in under the pre-existing vaccine liability exemption cover. Another Biden scandal.

BTW, I am finding the newish Google AI search summaries to be very helpful on a host of otherwise complicated ‘sciency’ search questions—provided the query is carefully phrased. Real time saver, and easy to verify.

Curious George
Reply to  Rud Istvan
January 4, 2025 11:09 am

It depends on how reliable your numbers are. A joke from early days of COVID-19. A coroner dictates a death certificate: Cause of death: COVID-19. – But, Doctor, he has three deep stab wounds. – Add: With complications.

Unfortunately, since 2008 I don’t trust official communications much.

Scissor
Reply to  Curious George
January 4, 2025 11:40 am

Me neither.

Reply to  Curious George
January 4, 2025 3:52 pm

It took until 2008?

David Wojick
Reply to  Rud Istvan
January 4, 2025 11:41 am

Just the time saver I suggested it would be:
https://www.cfact.org/2024/09/18/ai-could-take-your-computer-from-search-to-research/
Thanks Rud

Rud Istvan
Reply to  David Wojick
January 4, 2025 11:50 am

Prescient article.

Reply to  Rud Istvan
January 4, 2025 12:39 pm
bobpjones
Reply to  Rud Istvan
January 4, 2025 2:28 pm

there is no even complicated possible connection between the gut and brain”

I beg to differ, when my gut has not had food for a few hours, the brain begins to fog, eyesight blurrs, ears buzz and mouth drools. lol

Reply to  Rud Istvan
January 4, 2025 8:35 pm

I forget which but it was day 2 or 3 after her birth when my granddaughter’s autism was observable. Of course I had no idea what it was, only that it was not typical infant behavior. Pressure from other family members eventually led to professional examination and evaluation well before 18 months.The intervening 19 years have shown some inaccuracies in the original diagnosis but no major faults.

c1ue
Reply to  Rud Istvan
January 5, 2025 8:31 am

I have found numerous errors from Google AI search summaries.

ferdberple
January 4, 2025 10:45 am

Vaccines are not cut and dried.

Vaccines have the power to reduce infection rates but they can also can work in reverse to increase infection rates. For this reason we lack vaccines for many diseases.

What is suspect is combined vaccinations. Because vaccines can operate on a knife edge we should not be surprised there is the potential for side effects with combined vaccines.

For example during the covid epidemic there were virtually zero cases of the flu. Did covid cure the flu.

Rich Borba
Reply to  Willis Eschenbach
January 6, 2025 2:11 pm

How many vaccines were included in the largest study on autism and vaccinations

Based on the search results provided, the largest study examining the potential link between vaccines and autism focused on a single vaccine: the MMR (measles, mumps, and rubella) vaccine[1][3]. 

This comprehensive Danish study, which tracked 657,461 children over a 20-year period, specifically investigated the MMR vaccine and found no link between it and autism spectrum disorder (ASD)[1]. The MMR vaccine protects against three conditions: measles, mumps, and rubella.

While other large-scale studies and meta-analyses have been conducted, they also primarily focused on the MMR vaccine. For instance, a study of over 95,000 children, including 15,000 unvaccinated 2 to 5-year-olds, specifically examined the MMR vaccine in relation to autism risk[3].

It’s worth noting that some studies have looked at other vaccine components, such as thimerosal, but these investigations were not described as being larger than the MMR studies in the provided search results[4].

In conclusion, the largest study on autism and vaccinations included one vaccine (MMR) that protects against three conditions.

Citations:
[1] https://www1.racgp.org.au/newsgp/clinical/new-research-again-confirms-no-link-between-mmr-va
[2] https://www.chop.edu/vaccine-education-center/vaccine-safety/vaccines-and-other-conditions/autism
[3] https://www.autismspeaks.org/science-news/no-mmr-autism-link-large-study-vaccinated-vs-unvaccinated-kids
[4] https://pmc.ncbi.nlm.nih.gov/articles/PMC2908388/
[5] https://www.npr.org/sections/health-shots/2019/03/04/699997613/a-large-study-provides-more-evidence-that-mmr-vaccines-dont-cause-autism

Reply to  ferdberple
January 4, 2025 1:23 pm

For example during the covid epidemic there were virtually zero cases of the flu. Did covid cure the flu.

The same precautions that slowed the spread of Covid (lockdowns, social distancing and, yes, facemasks for spittle) also prevented the spread of Flu.

If they didn’t they wouldn’t have worked.

Derg
Reply to  MCourtney
January 4, 2025 1:51 pm

Wasn’t there an island who only allowed vaccinated people to enter and Covid still arrived on the island. Maybe those islanders should have worn masks and social distanced 😉

MarkW
Reply to  Derg
January 4, 2025 4:24 pm

Being vaccinated doesn’t mean you can’t also be a carrier. Especially if the vaccination is recent.

Jeff Alberts
Reply to  MCourtney
January 4, 2025 5:49 pm

yes, facemasks for spittle”

Spittle clearly wasn’t the problem.

Reply to  Jeff Alberts
January 5, 2025 8:47 am

Don’t be ridiculous. Phlegm sneezed over the buffet table is a problem. My mother taught me that.

Richard Greene
Reply to  ferdberple
January 4, 2025 2:28 pm

In an average year, 36,000 Americans die from the flu based on computer models, NOT death certificates. In 2020, US all cause mortality increased by 535,000 (Census data) — up 19% from 2019. 351,00 deaths were blamed on Covid which was the only known reason for such a huge mortality increase.

The 351,000 may be wrong but it was a conservative estimate. So what if EVERY flu related death was misdiagnosed as a Covid death? That would not make much difference in the obvious conclusion. Covid killed a lot of elderly Americans prematurely in 2020.

Tom Johnson
Reply to  Richard Greene
January 4, 2025 7:32 pm

Death certificates aren’t always accurate. A number of years ago a study of death certificates in California indicated that left-handed people died at a much younger age than those who were right-handed. 15 years younger was noted. The study went on to speculate that such items as scissors, power saws, and even coffee mugs and letter openers which are mostly designed for right hand use might be a factor. It was later pointed out that the study simply proved that people tend to die after they grow old. The generations that were dying in the study were born at a time that children were actively discouraged from using their left hands. Those that were dying were mostly right-handed and the only left-handed decedents were much younger.

Covid deaths likely had societal factors, as well. Prominent being federal payments to hospitals and doctors for treating Covid patients. There was a strong financial incentive for that to be the listed cause of death.

Derg
Reply to  Tom Johnson
January 5, 2025 3:03 am

Bingo, gun shot death in CO was a Covid death.

JulesFL
Reply to  Richard Greene
January 5, 2025 5:53 am

In the UK, average age of death by Rona exceeded, thereabouts, average lifespan.

January 4, 2025 10:51 am

Finally, some folks say that vaccinations don’t really protect us against diseases. While this may well be true about the COVID-19 vaccine, …
_______________________________________________________________________________

Not only does the COVID vaccine not provide full protection, it has some nasty side effects. Besides myself, I know too many people that have been vaccinated and came down with COVID more than once.

And from personal experience, there’s this from a very short Google search:

PMC PubMed Central®
Reviewing the official vigilance database, we found that heart rhythm disorders after COVID vaccination are not uncommon and deserve more clinical and scientific attention.

Reply to  Steve Case
January 4, 2025 1:31 pm

The Covid mRNA shots are mistermed as vaccines. The jabs from Pfizer, Moderna and the others meet all the elements defining a gene therapy treatment, and none of the elements defining a vaccine: most notably that a vaccine prevents a virus from making you ill or a transmitter. AFAIK there is only one Covid vaccine, which is Novavax coming on the scene later because developing a real vaccine takes more time than experimental mRNA. Thus this is a more complicated story to sort.

https://rclutz.com/2023/08/23/novavax-the-only-real-covid-vaccine/

Martin Brumby
Reply to  Ron Clutz
January 5, 2025 8:08 am

Worth mentioning that here in UK, the Astra Zenica shot was a very big deal, “UK leads World” and all that bunker.

I think it took less than a month for Norway, Sweden etc. to point out alarming number of blood clot problems and suspending this “vaccine”

Now it has been banned almost everywhere and can’t be obtained even in the UK for love nor money.

Didn’t stop the geniuses who developed it from getting Knighthoods and all the rest.

Numerous deaths specifically blamed by official Coroners on Astra Zenica. Yellow Cards almost without number.

No politician has apologised and the politician who highlighted problems and very popular in his constituency (Andrew Bridgen) saw his share of the vote six months ago drop mysteriously from over 60% to less than 5%. Odd.

I have several friends who have received eight or more booster shots of various Covid shots. They still keep getting “Covid” diagnoses.

Neither safe, nor effective.

January 4, 2025 11:23 am

Interesting, I was researching casein in relation to some gut issues I was having with protein supplements, one that was supposed to be lactose free. I am a pancreatic cancer patient so eating and maintaining weight is getting complicated.
As a result of that research I found a reference to a connection between casein, gluten and autism, suggesting that this might be an autism diagnostic tool.

A friend in an email conversation had mentioned perplexity.ai not an hour ago. When I asked it about the connection it confirmed what I had read less than a week ago.

Cool tool!

hdhoese
January 4, 2025 11:30 am

I remember the horrible polio epidemic quite well, two in my high school class died from it and iron lungs were the stories of many days. I seem to remember once reading about an older failed vaccine attempt maybe in the old world. While successful examples abound how important is history except for checking all those carefully? The live vaccine did also have some later minor effects, viruses being the nasty critters we deal with. It would be interesting to search many subjects, but there still seem to be problems despite the intelligence claim with numbers clearly important to increasing the necessary information. Is this just another exponential increase? It seems important as you said about no one question. Examples just received.
https://www.medscape.com/viewarticle/respiratory-syncytial-virus-vaccine-protects-older-adults-2025a100000j
https://www.newscientist.com/article/2462356-ai-chatbots-fail-to-diagnose-patients-by-talking-with-them/

This one not directly medical bothered me even before I learned where it came from. Raised questionable (unnecessary)centralization again? Journals spreading like viruses? International Journal of Data Warehousing and Mining
https://www.devdiscourse.com/article/education/3201363-big-data-in-preschool-education-bridging-gaps-and-enhancing-safety
https://www.igi-global.com/pdf.aspx?tid=361891&ptid=333260&ctid=4&oa=true&isxn=9798369324622

Fran
January 4, 2025 11:34 am

I actually saw one of the last smallpox epidemics in India in the early 1960’s. The vaccinator (an official position) had been lazy. The vaccine was miraculous in stopping the spread. One 11 y/o girl I saw on her death bed was kept back because grandma said she would be better able to look after 2 younger brothers when they were sick after the vaccination. People recognized the benefits and there was NO “antivax” talk in the villages – unlike the recent mRNA preparation, its efficacy was obvious.

That said, Kennedy is right to question the lack of proper testing – all cause mortality assessment is key. By and large, I agree with Willis as to the efficacy of vaccines against the specific diseases. But having spent my life working with drugs of all sorts, I refused the mRNA prep on the basis of not enough study.

A Danish group has been doing this in Gabon (?). They have been looking at all cause mortality (ACM) after vaccination for diseases that are not present (ie, the effects of the vaccine independent of its specific effects on the disease in question). Live vaccines reduced ACM over the next 5 years. Subunit and dead vaccines increased it. They propose that the schedule of vaccines be altered such that the beneficial type is given last in the series (eg, measles).

The WHO was finally goaded into checking into this and did so by not including children who died in the study!

https://www.nature.com/articles/s41577-020-0285-6

One of the most toxic childhood vaccines was for pertussis. That was changed to a subunit (ie synthesized single antigen) some years ago. The result is that efficacy is dropping as pertussis slowly evolves.

Fran
Reply to  Fran
January 4, 2025 11:42 am

PS – As an old fashioned stick in the mud, I will continue to cruise the literature to find out about things. For example, years before any nasty side effects of statins were acknowledged, I found a bunch of case reports and small studies in obscure journals. In regard to ACM, statins don’t change it – you just die of something else.

Rud Istvan
Reply to  Fran
January 4, 2025 12:47 pm

Both my Patricia and two of our acquaintances in our complex had severe adverse reactions to statins. Happens in about 4% of cases.

Admin
January 4, 2025 11:48 am

Willis, you horribly conflate the issues being discussed today with those being discussed a decade and a half ago, most noticeably Wakefield and the MMR vaccine.

That is not the issue people are currently concerned about. So toss every MMR study you note as irrelevant. What people are concerned about is what is noted in the image below. Toss every study on non-US residents, because we are literally the only country doing this to our children.

Your charts about the effectiveness of vaccines conflate appropriate use and administration as adults and teens where there are trackable benefits with the mass dosing of infants, which is the issue people are discussing.

When I said on X that much of this has no medical benefit I was talking about infants being given vaccinations only appropriate for adults.

This level of over medicating can cause all sorts of allergic reactions with unknown effects and parents have noted thousands of anecdotal associations with the MODERN VACCINE SCHEDULE.

Stop arguing as if this was 20 years ago.

The study I noted in NoCal, indicates an association. It is valid to study further.

And the fact that Covid is on the dosing schedule for infants should make you want to run and get torches to burn down the CDC.

The current childhood vaccination schedule, being exempt from any liability, is an obscene conveyor belt of Federal funds being handed to pharmaceutical companies who will keep putting more and more of these abominations on the childhood schedule.

GdkVyq4XoAAVuz5
Reply to  Willis Eschenbach
January 4, 2025 2:21 pm

The studies you use as refutation are all irrelevant to the issue at hand.

The illustrations of effectiveness you use are irrelevant to the issue at hand.

Sorry I skimmed through, but the issues at hand are what matters.

Complete regulatory capture of the NIH, CDC and FDA by pharmaceutical companies and obscene amounts of money being made pumping inappropriate crap into infants.

Reply to  Willis Eschenbach
January 4, 2025 8:51 pm

Why were the vaccine/autism studies limited to the MMR vaccine? Did no one think that the increasing frequent diagnosis of autism might be related to some other?
Yes, I am aware that the increase could be due to a greater awareness of symptoms.

Reply to  AndyHce
January 4, 2025 9:25 pm

Wakefield.

All the MMR studies were to refute Wakefield.

Richard Greene
Reply to  Charles Rotter
January 4, 2025 2:39 pm

That CR comment is Best of The Thread.

I wish there was a vaccine that prevents leftism, which I consider to be a mental disease.

I encourage an ADMIN vs. AUTHOR verbal war for my entertainment.
Snide remarks. Character attacks. Comparison of family members to various farm animals and/or their digestive waste products. And then, for a finale, bring in comparisons to Hitler

Reply to  Richard Greene
January 4, 2025 3:17 pm

We’re flattered you think our admin vs. author feud could be as entertaining as watching paint dry.

But really, escalating to farm animal analogies and Hitler references? That’s like bringing a squirt gun to a flamethrower fight.

We prefer our battles to have a bit more class, or at least not smell like the aftermath of a barnyard riot. But thanks for the entertainment blueprint—your comment is a goldmine for our next ‘How Not to Engage Online’ seminar. Maybe stick to suggesting we compare notes on the weather next time; it’s less… fertilizer-y.

Reply to  Richard Greene
January 4, 2025 5:34 pm

I wish there was a vaccine that prevents leftism”

Yet you espouse so many leftist ideology and fallacies… Go figure !!

Richard Greene
Reply to  bnice2000
January 4, 2025 11:43 pm

To a science denier like BeNasty, claiming CO2 is a greenhouse gas is leftism. Thanks for BeNasty Dumb Comment #128

Derg
Reply to  Richard Greene
January 5, 2025 3:06 am

Why are you a Holocaust denier 😉

Izaak Walton
Reply to  Charles Rotter
January 4, 2025 4:48 pm

putting more and more of these abominations on the childhood schedule.” Of course the preferred option should be to let kids die rather than being vaccinated. Kids are exposed to orders of magnitude more viruses and bacteria every day than vaccinations and if their immune system can cope with that then there is no issue with the vaccinations.

Reply to  Izaak Walton
January 4, 2025 5:29 pm

Barking seal talking point

were-saving-lives
Izaak Walton
Reply to  Charles Rotter
January 4, 2025 8:00 pm

Exactly. Vaccines save lives. The more the better.

Reply to  Izaak Walton
January 4, 2025 5:37 pm

Kids are exposed to orders of magnitude more viruses and bacteria every day”

Yes, it is what builds up immunity for most diseases.

It is only the major ones we should be trying to eradicate.

Too many vaccines slows the natural immunity development.

Izaak Walton
Reply to  bnice2000
January 4, 2025 7:03 pm

Nonsense. Vaccines do not slow the natural immunity but the exact opposite. They speed it up by exposing people to weaken variants before their immune system meets the real thing.

Derg
Reply to  Izaak Walton
January 5, 2025 3:08 am

Hopefully you are boosted. If you think all vaccines are safe and effective you have not paid attention to the ones pulled from the market 🙁

MarkW
Reply to  Derg
January 5, 2025 11:36 am

There are a lot of foods that are pulled from the market every year because of contamination or other reasons. Yet most people will argue that our food supplies are safe to eat.

Derg
Reply to  MarkW
January 5, 2025 1:34 pm

They put those foods back on, when drugs are pulled for good.

Reply to  Izaak Walton
January 5, 2025 7:00 pm

Vaccines are very useful technology, and using a number of them to control severe diseases has been of great benefit. But it is not necessarily a straight-line relationship Izaak. The immune system is impressive, but if you overload it with unnecessary challenges of minor ailments, you can reduce the response to the more severe pathogens out there.

And if you challenge it with the wrong antigens, you can actually have a negative effect on the immune response: now called Antibody-Dependent Enhancement (ADE):
https://www.nature.com/articles/nm0109-21

Respiratory syncytial virus (RSV) is the leading cause of lower respiratory tract infection in infants. By two years of age, more than 90% of children have been infected at least once with this virus. RSV infections typically cause mild illness; however, severe respiratory disease can occur and is associated with symptoms such as bronchiolitis and wheezing. 



During the 1960s, researchers thought they had a vaccine. But they were wrong—the experimental, formalin-inactivated RSV vaccine administered to children not only failed protect them but also exacerbated their illness in response to natural RSV exposure. Many of the children developed bronchiolitis and pneumonia so severe that it required hospitalization, and, unfortunately, two of the vaccinated children died. 

Vaccines have been linked to hypersensitivity reactions (fairly rate, but it certainly happens) so it seems logical they may be connected to allergies, but that is as yet unproven.

Reply to  Charles Rotter
January 5, 2025 6:43 pm

Nice study here:

Well thought out, comparing vaccinated vs unvaccinated homeschooled children removed most confounding issues.

Childhood vaccines home-school study Hooker and Miller

Analysis of health outcomes in vaccinated and unvaccinated children: Developmental delays, asthma, ear infections and gastrointestinal disorders
Brian S Hooker and Neil Z Miller

https://pubmed.ncbi.nlm.nih.gov/32537156/

Conclusion: In this study, which only allowed for the calculation of unadjusted observational associations, higher ORs were observed within the vaccinated versus unvaccinated group for developmental delays, asthma and ear infections. 

Results: Vaccination before 1 year of age was associated with increased odds of developmental delays (OR = 2.18, 95% CI 1.47–3.24), asthma (OR = 4.49, 95% CI 2.04–9.88) and ear infections (OR = 2.13, 95% CI 1.63–2.78).

In a quartile analysis, subjects were grouped by number of vaccine doses received in the first year of life. Higher odds ratios were observed in Quartiles 3 and 4 (where more vaccine doses were received) for all four health conditions considered, as compared to Quartile 1.

In a temporal analysis, developmental delays showed a linear increase as the age cut-offs increased from 6 to 12 to 18 to 24 months of age (ORs = 1.95, 2.18, 2.92 and 3.51, respectively). 

Reply to  Willis Eschenbach
January 5, 2025 10:20 pm

I’m very pro-vaccine, and it was only a nasty Covid vaccine reaction and the stonewalling which followed that sent me on a questioning path recently.

And I hadn’t realized the prescribed childhood vaccine list had ballooned to the extent it has, or that in the US the pharma groups have successfully lobbied to be protected against litigation by Congress (and possibly initially with fair intent, but of course they are can now put forward safely a greatly extended list of ‘life-saving’ vaccines with a reasonable amount of impunity). https://www.congress.gov/bill/99th-congress/house-bill/5546

On the study:

Certainly fair comments from an AI source point of view, and long lists always look impressive, but practically, this was an observational study, not a lab experiment.

This sort of study must be observational, randomly allocating prescribed vaccines to treatment and control groups in humans is a bit difficult.

Yes, it will also always be anonymized data: compounding factors would not be recorded (socio-eoconomic status, parental education and medical history)

And it is quite true that 3 practices may not represent the broader population. Repeated replication of the study would be necessary before anyone got too alarmed.

But, questions are raised, and further studies are warranted.
Asthma cases in vaccinates were increased 6x, developmental delay doubled, ear infections increased 50%.

A similar study with similar outcomes is here: albeit with smaller numbers.

Pilot comparative study on the health of vaccinated and unvaccinated 6- to 12-year-old U.S. children.Anthony R Mawson, Brian D Ray, Azad R Bhuiyan and Binu Jacob 2017
https://www.oatext.com/pdf/JTS-3-186.pdf

January 4, 2025 11:49 am

What I’ve presented above is how to use perplexity or any AI to unearth scientific evidence about very specific questions of a) autism and vaccines and b) the overall efficiency of each vaccine.

AI is most problematic if you ask it for specifics such as the names of specific papers. It’s kinda like if I asked you for specifics then you’d have to wrack your brain for the name and year and so on. Models don’t include specific information like a database does, they have memories of what they were trained with and it’s very much like your own memories.

Having said that, AI can look things up on the internet if it’s built to do so and I haven’t looked into perplexity.

It’s definitely worth confirming facts returned by AI if they’re key to its argument.

January 4, 2025 11:56 am

Here is an answer from Perplexity to a climate-related question. I thought the response was pretty good.
****************

Me: Please identify a scientific study determining the reliability of global surface air temperature projections by propagation of error….

Answer
A scientific study determining the reliability of global surface air temperature projections by propagation of error was conducted by Patrick Frank and published in 2019 in the journal Frontiers in Earth Science. The study is titled “Propagation of Error and the Reliability of Global Air Temperature Projections”1.Key FindingsThe study’s main conclusions include:

  1. General Circulation Model (GCM) air temperature projections are linear extrapolations of greenhouse gas forcing1.
  2. These linear projections are subject to linear propagation of error1.
  3. The study identified a significant calibration error in CMIP5 climate models, specifically an annual average ±12.1% error in global annual average cloud fraction1.
  4. This error results in a ±4 Wm^-2 uncertainty in long-wave cloud forcing (LWCF), which is much larger than the annual average change in tropospheric thermal energy flux from increasing greenhouse gas forcing since 19791.
  5. When propagating this error through various climate projections, including IPCC scenarios, the study uncovered a ±15°C uncertainty in air temperature at the end of a centennial-scale projection1.

MethodologyThe study employed the following approach:

  1. Accurately reproduced GCM global air temperature projections using an emulation model.
  2. Demonstrated that advanced GCMs project global air temperature as a linear extrapolation of fractional greenhouse gas forcing.
  3. Propagated the identified calibration error through CMIP5 global air temperature projections to assess their physical reliability1.

ImplicationsThe author argues that this analysis reveals significant uncertainty in global air temperature projections, calling into question their reliability. The study emphasizes the importance of including physically valid error bars in temperature projections, rather than relying solely on model precision1.It’s worth noting that this study has generated some controversy and debate within the scientific community. Other researchers have questioned aspects of the methodology and conclusions, highlighting the complex nature of climate modeling and projection reliability assessment2.

Richard Greene
Reply to  David Dibbell
January 4, 2025 11:56 pm

All that AI claptrap and you still do not know the likely margin of error for global average temperature statistics.

 Pat Frank argues that there is an uncertainty in the cloud forcing. And he thinks he knows how much. He does not. There are no accurate measurements of cloud effects on the climate. There is only a proxy of percentage cloud cover with at least a +/- 10% margin of error for a global average. Changes in the past few ecaes were claimed to be -7% which is not likely to be statistically significant.

Actual effects of clouds require data for the type of clouds, height of clouds and timing of clouds for several decades, at least. Only then could a trend of global average cloud effects be calculated. Pat Frank seems clueless about the limited knowledge of the climate effects of clouds.

Reply to  Richard Greene
January 5, 2025 2:07 am

Nice to hear from you, Richard! I hope you are well.

Pat Frank argues that there is an uncertainty in the cloud forcing. And he thinks he knows how much.” No. Read and understand the paper. He simply begins with published values from others to perform his analysis.
.
“Pat Frank seems clueless about the limited knowledge of the climate effects of clouds.” No. His paper is not about clouds themselves, nor the influence on the climate trends. It’s about the application of calibration error to the step-iterated outputs of the GCM “instrument” to estimate the uncertainty and therefore the reliability of the resulting projections.

“All that AI claptrap…” I am under no illusions about what these LLMs do. But in this case Perplexity did a better job of summarizing the paper than you have. You have completely missed the point of Pat Frank’s analysis. You should do better.

One more point that connects the author of this WUWT post (Willis Eschenbach) and Pat Frank. Separately, they each demonstrated the successful emulation of the GCM global surface air temperature projections by simplified computation from the published “forcings.” The implications of this finding are widely underappreciated even here at WUWT.

Again, be well. We agree on many points. But you are really “missing the boat” on this one.

January 4, 2025 11:58 am

Willis, I question the accuracy of the graph showing the introduction of the polio vaccine in 1957. My vague recollection is that I was lined up along with the rest of my young classmates for my sugar cube about 5-years earlier, perhaps about 1951. My recollection is that it was at a small elementary school in McHenry, Illinois. In fact, your graph suggests that polio deaths actually started to decline in the early-1950s; although, with only decade dates on the x-axis, it is difficult to determine the year for the peak in cases. By 1957 I was 15 and a sophomore in high school in the SF Bay Area. My recollection of things is pretty good from that point on (Until more recently! 🙂 ). Thus, I think that I would remember if I was given the sugar cube at that age. I also vaguely remember being given a sugar cube on more than one occasion.

Nick Stokes
Reply to  Clyde Spencer
January 4, 2025 2:12 pm

Willis, graph is right, at least for the precipitous fall. I was surprised at the steepness of the prior rise. Your memory is muddled – Wiki explains:

“The first successful demonstration of a polio vaccine was by Hilary Koprowski in 1950, with a live attenuated virus which people drank.[10] The vaccine was not approved for use in the United States, but was used successfully elsewhere.[10] The success of an inactivated (killed) polio vaccine, developed by Jonas Salk, was announced in 1955.[2][11] Another attenuated live oral polio vaccine was developed by Albert Sabin and came into commercial use in 1961.”

Salk was the first mass vaccine, starting after 1955.

Reply to  Nick Stokes
January 4, 2025 5:23 pm

My father had polio as a child and had a crippled leg for the rest of his life. Growing up in the UK through the 50s everyone was terrified of polio outbreaks. When the vaccine was introduced in 1955 (Salk) everyone from our primary school went to get the shot. Through my teens (60s) I helped my parents raise funds for the British Polio Organization to support the international elimination of polio worldwide. I can’t imagine anyone who lived through the 50s being opposed to polio vaccines!

Reply to  Nick Stokes
January 4, 2025 9:36 pm

It certainly is a possibility that my memory is muddled. I may be confusing the event with my smallpox vaccination. Unfortunately, at my age, both of my parents are dead so I can’t ask them if they remember when they had to sign the permission slips.

Perhaps you would be good enough to speculate on why it appears that both the number of polio cases and deaths seem to start a decline about 1952 on the graph, certainly well before 1957. I believe that (based on counting back from the dot at 1960) WE has placed the arrow for 1957 incorrectly. ’58-’57 appears to be a doublet, albeit ambiguous. It appears to me that WE’s arrow actually probably corresponds to 1954, not 1957, and the peak in cases occurred in 1952. The administration of the sugar cubes would have started after the September start of school (after the US Labor Day) and before November snows.

I don’t believe everything I read in Wiki’.

Reply to  Clyde Spencer
January 5, 2025 8:59 am

You appear to think that polio occurred at a constant level, actually there were ‘outbreaks’ which were a spike followed by a drop in cases followed eventually by another spike. The widespread introduction of the vaccine in 1955 broke this cycle, effectively eliminating the disease. As far as I recall the Salk vaccine was an injection and the sugar cube was only used for the Sabin vaccine which was introduced later, first used in US in 1960.

Reply to  Phil.
January 7, 2025 10:36 am

I made no assumptions about the character of the annual infection rates. I was referencing the graph supplied by Willis.

I do remember getting a sugar cube while in school. (At least once, if not twice, but then my parents moved from Illinois, to Arizona, to southern California and then Northern California during this time, 1952-1960. By the time I graduated from the 12th grade, I had attend more than 12 different schools from the US Midwest to the West Coast. I may have received more than one cube out of an abundance of caution by the schools.)

According to Wikipedia, “In 1954, the year leading up to the [safety] announcement, polio was killing more American children than any other infectious disease. Jonas Salk‘s vaccine was made ready for its third and final field tests. It became the most elaborate program of its kind in history, involving 20,000 physicians and public health officers, 64,000 school personnel, and 220,000 volunteers.” That is, the announcement of the safety [April 1955], and subsequent approval, of the Salk vaccine came after two previous field trials. I may have been in one of those two previous field trials. To further bracket the time, “Six months before [October(?) 1954] Salk’s announcement [April 1955], optimism and hope were so widespread that the Polio Fund in the United States had already contracted to purchase enough of the Salk vaccine to immunize 9,000,000 children and pregnant women the following year. Therefore, by the time that school started in 1955, the polio immunization program was in full swing.

“As far as I recall the Salk vaccine was an injection and the sugar cube was only used for the Sabin vaccine which was introduced later, first used in US in 1960.”

I graduated from high school in June of 1960 and I’m certain that I received the vaccine well before I turned 18 in January of 1960! It is, after all, principally a childhood disease. I don’t recollect adults being inoculated routinely, as was done with school children.

I do remember getting numerous vaccinations with pneumatic injectors, in both arms, while in Basic Training in 1966. However, to the best of my knowledge, I never received a polio immunization injection.

rbcherba
January 4, 2025 12:05 pm

Willis, you should have asked Perplexity for the correlation between autism spectrum rates and government programs to support autism patients. Or asked why the number of children identified as “slow,” “intellectually handicapped,” and other terms related to low intellectual ability have virtually disappeared. Most of these children now benefit from being in the “autism spectrum.”

Rud Istvan
January 4, 2025 12:13 pm

Separate more general comment about childhood vaccines, having just done some quick research in a complicated area where RFKJr has expressed concerns. (I previously posted research that his stance on stopping fluoridated water is rational and well supported.) There seem to be three major concern components.

  1. The current AAP/CDC officially approved schedule, which is pretty intense as CtM indicated. The concern is sufficiently great that presently only 63% of parents follow the schedule. One result is an uptick in measles cases.
  2. Preservatives. The big one was thimerosal (a Mercury based compound), which has mostly been phased out. It is now formaldehyde, still used.
  3. Adjuvants, the most common of which are various aluminum salts.

IN my opinion, the medical community pro arguments are not very compelling, and it looks like there is insufficient research on these issues.

January 4, 2025 12:19 pm

Willis, I think that it is wise to suggest caution in interacting with LLMs. It has been my experience working with ChatGPT and Copilot, that they will make unreserved assertions that are simply wrong when I have asked questions about the climate. When challenged, they promptly back down, rather than arguing. Therefore, just because your experiment didn’t provide counter evidence, it doesn’t mean that ‘Persnickety’ isn’t hallucinating, or at least reflecting the bias of its training. Lack of evidence isn’t evidence against something. Indeed, with the track record for LLMs being what it is, I suspect that a little more research is warranted. Might I suggest some research not using LLMs, and if you find something, challenging ‘Persnickety’ with, “What about the study by ___?” It would be surprising if it would react differently than the other LLMs when challenged.

I think that your conclusion about the reliability of your source might be premature, and I say that as someone who is not an antivaxer and regularly gets recommended vaccinations without hesitation. I just think that the design of your experiment could use some tweaking.

Reply to  Clyde Spencer
January 4, 2025 12:38 pm

Therefore, just because your experiment didn’t provide counter evidence, it doesn’t mean that ‘Persnickety’ isn’t hallucinating, or at least reflecting the bias of its training.

It absolutely is reflecting its training. That’s the point isn’t it? And will hallucinate if you take it down a “story” path. AI will absolutely create a narrative of anti vaccination if you direct it to do so.

The specific questions you use and path the narrative takes is all important. If you force your own biases into the path then expect to end up with what you want to be told.

Sean2828
January 4, 2025 1:00 pm

Would you be willing to pose another question to your AI search?

List major studies showing that autism either is or is not associated with C-Section delivery of babies.

I’ve often wondered if that is a question the medical establishment does not want asked.

Rud Istvan
Reply to  Sean2828
January 4, 2025 1:12 pm

Don’t need AI to get a firm answer. In 2023, 32.4% of births were via C section—WHO says rate should be 10-15%. In 2023, 2.8% of kids were diagnosed with ASD by age 8. So, no significant relationship.
Twin studies show that between 60 and 90% of autism is heritable, so largely a matter of genetics.

Reply to  Rud Istvan
January 4, 2025 5:33 pm

Twin studies show that between 60 and 90% of autism is heritable, so largely a matter of genetics.

It seems more likely to me that the genetics of the mother whose inclination (or need) to have a c-section may be more highly correlated than anything related to the procedure of the c-section itself. I’d start there if I was investigating.

But people dont like to be personally responsible and would rather blame the procedure if there is indeed a causal link. Even though “responsibility” would be entirely blameless in the case of genetic causes.

MarkW
Reply to  Rud Istvan
January 5, 2025 11:43 am

One problem with trying to compare ASD rates and C section rates is that doctors don’t just randomly decide which women get C sections and which don’t.

In general C sections occur after a woman has been in labor for a long time, or if there is some sign of fetal distress.
In both cases, the baby has been put through more stress than they would have endured had the labor proceeded smoothly.

Reply to  MarkW
January 5, 2025 12:01 pm

In general C sections occur after a woman has been in labor for a long time, or if there is some sign of fetal distress.

I wouldn’t say that was the general case. Many women have c-sections because they don’t go into labor at the right time and plenty of women choose to have a c-section at a convenient time or because they simply don’t want to go through childbirth and they have it scheduled.

Reply to  TimTheToolMan
January 7, 2025 9:41 am

Breech position of the baby is one reason, it was in the case of my daughter.

JonasM
Reply to  MarkW
January 5, 2025 12:53 pm

My wife was in the hospital 2 weekends in a row for attempted labor induction. Neither time succeeded. The following Monday, C-section was performed. Our doctor’s first words were “Oh, that explains it”. My daughter’s head at birth was beautifully round, not the pointy shape needed to trigger mom’s nerves to induce labor. If it was 100 years ago, one or both would probably have died. In today’s world, my daughter is waiting on results for graduate school admission. A much better result. 🙂

Bob
January 4, 2025 1:10 pm

Very nice Willis.

January 4, 2025 1:21 pm

AI is changing how research is done. I don’t trust the AI algorithms for completeness.

Anyone interested in the vaccine safety debate should read the book below.

The answer you got Willis is wildly incomplete.

Turtles All The Way Down: Vaccine Science and Myth: Anonymous, O’Toole, Zoey, Holland J.D., Mary, Holland J.D., Mary: 9789655981049: Amazon.com: Books

Turtles All The Way Down: Vaccine Science and Myth Paperback – July 16, 2022
by Anonymous (Author), Zoey O’Toole (Editor), Mary Holland J.D. (Editor, Foreword)
4.8 4.8 out of 5 stars
   (1,289) 4.6 on Goodreads 480 ratings
See all formats and editions

If you are reading this, you are probably aware of the fierce debate surrounding vaccination and looking for information that will allow you to make the best decisions for yourself and your loved ones. Whether you are a parent or a parent to be, sorting through the many arguments on vaccines can be daunting. Still, you need an answer, a definitive one, to the crucial question: Who has it right in the great vaccine debate – the critics, who claim that vaccines often cause serious harm, or the medical establishment, which tells us that vaccines are safe and effective and the science is settled?

Rest assured, you have come to the right place. Turtles All the Way Down: Vaccine Science and Myth will resolve the vaccine question for you, once and for all. By the time you finish reading, not only will you see the answer clearly for yourself, you will also have the scientific references and specific quotes at your disposal that prove it – more than 1,200 of them – all from mainstream scientific papers and textbooks, the official publications of relevant government agencies, or manufacturers’ documents.

The book consolidates a great deal of information (accompanied by detailed analysis) that is scattered in hundreds of medical articles, books, and websites. All discussion is presented in clear and easy-to-understand language, so no medical education is required. It presents several original concepts in addition to laying a robust scientific foundation for the more established ones.

Some of the fundamental vaccine safety issues covered in the book are:

  1. How is safety demonstrated before a new vaccine is licensed? What technique do vaccine manufacturers use in clinical trials to make vaccines appear safer than they actually are?
  2. What “last ditch” technique is employed when the above one cannot be, and what are its grave (and damning) ethical implications?
  3. What is the scientific foundation of the safety of vaccination, and what practical tools does this body of science provide physicians to anticipate, diagnose, and treat vaccine injury?
  4. What fundamental flaws are built into vaccine adverse events reporting systems, and how are these systems used (or misused) by health authorities to support their safety claims?
  5. What kinds of post-marketing vaccine studies are conducted, and how can they be manipulated by researchers to produce “favorable” outcomes?
  6. Why would researchers want to skew vaccine research, and how could skewed results be promulgated by the scientific community?
  7. Why would medical journals publish faulty vaccine science? What is the role of the famed “peer review” in this process?
  8. What are “the studies that will never be done” by the medical establishment and how long it has resisted doing them? (Hint: more than 100 years!)
  9. What key CDC-recommended childhood vaccination guidelines were arbitrarily set, without an adequate scientific basis?

In addition, three cornerstones of vaccination lore are covered in depth:

  1. What is herd immunity, and how does it apply (or not) to the vaccines on the childhood schedule?
  2. What role did vaccines actually play in the historical decline of infectious disease?
  3. Was the paralysis associated with polio actually caused by the poliovirus? Is there a better explanation for the great paralysis epidemics of the 20th century? What are the “19 polio mysteries”?

The book is intended for parents overwhelmed by conflicting messaging on this important topic, but it is also an excellent reference for medical researchers and professionals who seek a better understanding of vaccine safety science. Whether you are new to the vaccine debate or a “veteran” seeking a deeper grasp of the science, this book is a must-read. It also serves as an excellent primer on vaccination to share with friends and relatives who may benefit from a deep dive into the subject.

4 Eyes
Reply to  Nelson
January 4, 2025 5:14 pm

Nelson, you have asked a whole lot of questions that could be directed to AI’s Perplexity. When we discuss things face to face with other people we rephrase questions if we feel the original question was not quite what we intended. After an hour or 2 (and a coupla beers) we often find we’re both getting closer to a reasonable understanding.

Reply to  Willis Eschenbach
January 4, 2025 8:31 pm

It is the custom in science, that when a person makes an assertion, the onus of providing support for that assertion falls on the person making the claim. You have largely met that responsibility by reporting your experience. However, the Scientific Method does not stop there. If, after publishing your conclusion(s), there are questions, objections, or countering interpretations and facts from your peers, the onus still falls on the original author to respond to those, not simply tell them to go do their own research. Otherwise, those who have raised legitimate concerns about the conclusion(s) are liable to assume that you can’t answer the questions, and — rightly so — dismiss your claims as being unsupportable. Perhaps interesting, but certainly not the last word on the subject.

Reply to  Willis Eschenbach
January 7, 2025 9:18 am

Clyde, this is exactly why I ask people to quote what they are referring to. I have no clue what your interjection refers to. It seems you think someone should be responding to something, but you’ve given no indication who the someone is and what the something is.

Pass.

Willis, it is obvious that you know “who the someone is” because you responded to me. I was responding to your comment at January 4, 2025 1:43 pm, immediately above my comment. It seems a waste of time and bandwidth to quote your entire comment that immediately precedes my comment. However, I have complied with your request with this comment.

Further, it seems that you expect people to abide by your requests but don’t do the same to clarify what you are responding to. “What is good for the goose is good for the gander.”

Dodgy Geezer
January 4, 2025 2:10 pm

My understanding is that AI services work by using a Large Language Module – a piece of code which is fed with gramatical rules and can provide coherent sentences. This is then ‘trained’ by feeding it with data. Many current ones simply use the entire Web for training.

So what you get is an average of the data provided. Great if you do not want anything controversial. Of limited use if you are exploring new boundaries.

Incidentally, I have looked at the Wakefield controversy. I understand that his paper was finally withdrawn, not because the data was false, but because of claimed irregularities in his obtaining consent from subjects. I hold no brief for him, but the whole episode looked like an establishment shake-down to me.

Reply to  Dodgy Geezer
January 5, 2025 1:05 am

There is another sort of AI altogether, machine learning. And this is really astounding. The book “Game Changer” by Matthew Sadler and Natasha Regan covers the chess games played by AlphaZero against Stockfish. You have to be a strong player to appreciate its play – so to get a feel, if you are not a player, find someone who is 2000 rated or higher and ask them about some of the key games.

People say this is not intelligence, it just looks to us like it. It looks to us like insight when its actually just an algorithm. Maybe. But it is certainly passing the Turing Test. Play through the games without knowing, and you would find them impossible to distinguish from play by a super strong grandmaster.

I have the feeling that machine learning and expert systems is what will change the world – is changing the world. Far more than LLMs.

Richard Greene
January 4, 2025 2:14 pm

In recent decades. the broader definition of autism has led to an increase in the number of people diagnosed with ASD. However, some of this increase may be due to improved detection and awareness, rather than a true increase in the incidence of autism. 

Parents are very reluctant to admit they have a child with below average intelligence or a troublesome, overactive child. The result is wanting to have their child diagnosed as a victim of autism or attention deficit disorder. When I was young, we were more honest. We called these children retards and juvenile delinquents. They were not seen as victims, their parents and the community were the victims!

The current practice of a huge number of vaccinations for children has never been proven to be safe. The CDC recommends vaccines before age two to protect children from 14 diseases. Children typically receive more than 50 shots before they reach adulthood

The same organizations that promote these vaccinations include Covid19 vaccines for children. Such vaccines are FAR more dangerous for children than Covid itself. Therefore, the organizations that promote the huge numbers of vaccines for children should not be trusted.

NOTE: I have not had any vaccinations in about 65 years and do not intend to get any vaccinations for the rest of my life.

I know that global warming has been pleasant in SE Michigan and other northern states for the past 50 years. But Google AI will claim global warming is dangerous because that is the leftist and government consensus position.

Editor
January 4, 2025 2:20 pm

w. ==> “I found it fascinating to see its judgment of the validity of the information it is examining” Perplexity did not evaluate the “validity” of the studies, only that they existed in the real literature and were not “made up”/”hallucinated” citations.

Perplexity is simply reiterating what is “most often said” in the material it has been trained. There is a slight tendency towards biasing depending on the “reliable source” function — such as “Newspapers of Record are reliable sources” “The IPCC is a reliable source” “The Lancet is a reliable source”.

Perplextity is a Large Language Model and can ONLY predict what the next most likely word should be based on the probability from its training. This means that it will say whatever has been said the most times in its training material. It has no way to know if the information is true or false.

Perplexity also can not look at the evidence — it can only look at what was said about the evidence.

So, you have discovered the “general consensus” found in the literature, on the ‘net, and in the press. But there is not necessarily any truth value in that (see Climate Change).

January 4, 2025 2:57 pm

I have no sway on X but will ask for your reinstatement.

Izaak Walton
January 4, 2025 3:17 pm

Willis,
why is the increase in the number of vaccinations any reason to be concerned? The alternative would be to let the kids catch the actual disease which is always going to be
a worse option. The immune system is capable of producing 10^18 different antibodies
and kids are exposed to orders of magnitude more bacteria and viruses every day than
that their immune system deals with compared with the tiny number of vaccinations that
they receive.

ferdberple
January 4, 2025 4:40 pm

Why is the condom so popular? From ChatGPT.

Here are the 10 most dangerous STDs with indications on whether an effective vaccine exists:

1. HIV/AIDS – No vaccine

2. Syphilis (Treponema pallidum) – No vaccine

3. Gonorrhea (Neisseria gonorrhoeae) – No vaccine

4. Chlamydia (Chlamydia trachomatis) – No vaccine

5. Human Papillomavirus (HPV) – Yes (HPV vaccine)

6. Hepatitis B – Yes (Hepatitis B vaccine)

7. Hepatitis C – No vaccine

8. Herpes Simplex Virus (HSV) – No vaccine

9. Trichomoniasis (Trichomonas vaginalis) – No vaccine

10. Pelvic Inflammatory Disease (PID) – No vaccine (a complication, not a disease itself)

The vaccines for HPV and Hepatitis B are effective in preventing infection and related complications, while others remain without vaccines due to challenges in developing one.

One of the more popular human pastimes and no vaccine is available for 8 of the 10 most dangerous infectious diseases that can result. The jury is still out on the dangers of acute lead poisoning.

Reply to  ferdberple
January 4, 2025 8:45 pm

It has been my experience that condoms are not actually popular. They are, instead, a pragmatic choice begrudgingly made to avoid something akin to “buyers remorse,” and given a choice of effective alternative actions, would rank low on the popularity list.

That points out one of the weak points of LLM AI programs. They don’t provide their definition of a key word in the question, or ask you to define what you mean by a commonly used word like “popular.”

Reply to  ferdberple
January 4, 2025 10:02 pm

If you have the bucks, there are very effective cures for Hepatitis C.

ferdberple
January 4, 2025 5:14 pm

I asked chatGPT about vaccine effects on our coevolved residents.

Answer:
Yes, human beings are host to a vast amount of foreign DNA, with estimates suggesting that microbial DNA (bacteria, viruses, fungi, etc.) outnumbers our own by a significant factor, and some studies suggest that up to 90% of the cells in our body are microbial.

As for widespread multi-species vaccines, there is potential for unknown risks, particularly when it comes to the microbiome and immune responses. Our immune system has evolved to balance interactions with various pathogens, commensals, and parasites, and disrupting this balance through vaccination or other means could have unforeseen effects. For example, helminth (worm) infections have been studied for their association with allergies and autoimmune diseases. Some research suggests that the lack of these infections in modern, sanitized environments may contribute to the increase in autoimmune diseases and allergies. This phenomenon is part of the hygiene hypothesis, which posits that reduced exposure to pathogens in childhood can lead to immune system dysregulation.

Therefore, while vaccines provide essential protection against harmful diseases, introducing vaccines that affect not just one pathogen but multiple species could alter immune system behavior in complex ways, potentially leading to autoimmune diseases or allergies in susceptible individuals. Research is ongoing to fully understand these intricate relationships.

Star Trek: The Way to Eden.