https://www.youtube.com/watch?...
About ten years ago a group developed a material with a magneto-restrictive property greater than the GMR Giant Magneto-restrictive material used in spinning disk drive heads. They used the modifier "Ludicrous" for the MR effect and submitted the paper to Nature. The paper got accepted, but they could not call it LMR. Too bad.....

A fine example of the nonsense some people will wilfully believe without any supporting data. Occult Chemistry
It was even reviewed in 2013 almost 100 years after being published.
Not sure if this is the type of content they expect from old books.

Or if too long for a tee shirt, print in a book like "Cracking DES."

“For the things we have to learn before we can do them, we learn by doing them.”
  Aristotle, The Nicomachean Ethics, circa 330 BCE

Pasteurization also reduces or eliminates zoonotic transmission
https://www.sciencedirect.com/...
https://www.popsci.com/health/...

I agree that most of the LLMs, including DeepSeek-R1, are not very good at generating functional code let alone performant code.
I suggest you try your coding tests with the ChatGPT 03-mini-high model. I have been testing LLMs for esoteric knowledge and code generation ability since 2023.
I have been mostly underwhelmed; however, the 03-mini-high can produce correct code that is more performant than some human generated code in my tests.

Go to this site and pick your favorite language (there are dozens to choose from):
https://rosettacode.org/wiki/E...

Run the code with timers to get an example of Human created code which may or may not be optimized.
Then go to chatGPT and pick the o3-mini-high model.
https://chatgpt.com/?temporary...
Give it this prompt:
        Create a program in the XX language that will find a counter example for Euler's sum of powers conjecture for a power of five and integers up to 250.
        Add timing code to report time to find first counter example.
        Optimize loops for speed.
Compare the speed of the generated code with the human generated code.
I have done this with Python and Julia.

Indeed, if you ask chatGPT to write obfuscated, unfathomable, self-modifying code it says that it can't do that as it does not advance computer science and understanding of programming. Or some similar nonsense like that. It already makes stuff up, er I mean hallucinates, so the attempts on controlling output are already a fail. Part of the problem is that it many people will accept the written word as fact if it is presented that way. It may be part of human nature and this tendency goes back millennia and concepts written by the ancient Greeks were accepted for as fact for a long, long time even when closer observation and a little critical thinking would have exposed the "facts" as nonsense. Take for example the word "bugonia", there is a long article on in in wikipedia.

You can think of American Foulbrood as anthrax for bees.
Indeed, the bacteria was classified as Bacillus larvae until 2006 when it was reclassified as Paenibacillus larvae.
The bacteria which causes anthrax is classified as Bacillus anthracis.
The bacteria form spores which are hard to kill and can be easily transported between hives.
Hence, the need to burn infected hives as they spores will always be there.
In Maryland, they used an ethylene oxide chamber instead of incineration of infected hives until recently.
The state bee inspectors changed the protocol as the chamber was hardly ever being used, i.e. little foulbrood in the state,
and ethylene oxide is now classified as a carcinogen and the residual gas left behind in the process was too great of a risk.

If I was keeping bees in an area with a high risk for American Foulbrood, I would use the vaccine in my colonies, as having to watch a hive being burned (with the bees in it mind you) would be too sad. Continually treating with antibiotics also has its risks.

To track them of course.
I put our logo on all of our bees so when a neighbor
complains that our bees are getting water out of their swimming pool,
I ask them if they can see our logo!

Search "bugonia" in Duckduckgo and the first result is the wikipedia page.

ChatGPT does not know the term "bugonia."
It does have information on statements made in the Wikipedia page such as Sampson's riddle in the Book of Judges.

If ChatGPT can't even respond with information in Wikipedia, how can it replace a search engine?

Federal funding agencies have tended to fund the work you have done rather than the work you propose in a grant application for a long, long time.
This makes it very difficult to investigate new ideas that may be outside of your published work by even a small amount.
True innovation is risky and if a project fails and there are no publications, the grant is considered a failure. Grant managers want
to show off their portfolio of grants with many publications.

There have been many initiatives to fund innovative research and younger investigators, but these grant application are reviewed
by the same people with the same ideas and prejudices on how applications should be reviewed.

Paracelsus (AKA Philippus Aureolus Theophrastus Bombastus von Hohenheim) was a physician, alchemist and philosopher in the early 16 th century.
He is considered to be the first chemist and toxicologist. His quote, "The dose makes the poison" is well known.
He thought that all medical practitioners should understand chemical principles and taught such principles to his medical students.
The medical establishment at the time did not fully agree with the need for understanding chemical principles, but chemistry prevailed and has been
a key part of medical student's education since that time.

Without understanding basic organic chemistry, it is not possible to fully understand biochemistry. I don't think that anyone would say that biochemistry is
not important for medical education. One can argue that with the proliferation of effective medications, a general practitioner only needs to understand
what medicine may alleviate what condition to be able to prescribe that medication and how it functions or is made is not important. However, there are
often cases where a patient has an adverse reaction and an understanding of how the drug works at the biochemical level may help to understand the
problem that the patient is having.

I have a Ph.D. in physical chemistry which takes an understanding of calculus and quantum mechanics.
I thought that some of the undergraduate organic chemistry classes were hard due to all of the named reactions and one
had to remember the names to answer problems on exams. Organic chemists like to refer to reactions by the discoverers name
rather than what it does. Memorization of names is hard for me which is why I found organic chemistry classes difficult.

If trees are growing faster and larger while growing weaker tissue this could lead to premature failure of the trees structure.