Paleonet: Mucking about with ChatGPT

Sat Oct 14 12:29:27 UTC 2023

Dear List-Members,
ChatGPT, although I never used it yet, but I tried, is another form of 
assembling information, analyzing it
and presenting it. It is an "evolutive succession"
of knowledge-availability.
You can, for example, ask "Bing" questions of the type:
Where is (name of a person, such as Luise ...)?
Words like "Where", "How", "When" etc. are understood by "Bing".
We had initially "data gathering" (geological maps, laboratory analyses, 
e.g. up to ca. 1970).
Then statistical analysis in addition, see the book, it was a bestseller, 
"Statistics and Data Analysis in Geology", from "Davis", I think from Kansa 
Geological Survey.
I mean the first edition, e.g. with the programs printed.
Then we had modelling.
The last generation of models requires extremely "comprehensive" data 
gathering before the start.
Parallel to that we had search engines like Altavista.
Later Yahoo.
Yahoo was a "manual" combing through information.
Google was, conceptually, AI, e.g. algorithms to
sort information according to relevance, however it is defined:
When in New Orleans the Hurricane Katrina happened,
e.g. the "ground-failure of the levees" Google had in
such artciles inserted Ads: "Cheap stay in New Orleans".
Later it got "profiling of users" with "inferred" interests: When I typed 
things for Mongolia,
I did not get geolgical things but always, from Google, "various types of 
touristic activities".
They merged the location (Muenster in Germany)
with age, and "what might the interest be?"
German T-Online is often "very bad": For me they displayed various type of 
funerals (earth, ocean, rest-forest Hummel in Germany, other), recently 
"cheap coffins" or it is "psychological warfare".
***
ChatGPT is the next step of "information-combing"
and displaying.
Comparable to my long-term habits of asking
questions with "When", "Where", "How much"
etc. with Bing.
Possible at Google they thought: We can do it better.
Whether they had been sucessful, is a question
of further developement. See search engines, I used them, really, like 
"Lynx" and "Gopher", e.g. early stages of the internet, compared to Google 
today.
***
For all here:
You can ask ChatGPT a queston like: How to set biological age back?
How long can we life science based?
Results should be a list of five papers:
Nature, 2011, Nature Medicine, Science, I think ca. 2014, New England 
Journal of Medicine, 2016.
With "eternal life is possible", even for seniors of today.
"2011" is adjusting a hormone 4-OHT, a "special testosteron" in the blood 
to ca. 1 ng/ml.
Hard science
Ageing and death is paleontologically very old.
That is: "Volvox stage", possibly ca. 2.5 Ga or
a little older, Early Proterozic or earlier.
Evolution very seldom "invents" something "fundamentally new".
In the standard case "evolutive inventoins" are "inherited" through 
evolution of "that line".
Implies: All animalia, at least all vertebrates,
at least all land-vertebrates, have the same type of "ageing".
Implies: Death question is solved, the contribution in "Nature, 2011".
***
Energy question is solved from me: Its published.
Some hints for the "most convenient way" I can give.
Means:
Deep wells, until the limit of plasticity, e.g. ca. 18
to 24.6 km in Germany, ca. 650 km in East Japan,
are "possible" and "cheap".
For New York too (until limit of plasticity) the well in the form of a 
spiral etc etc etc, double well, both connected, it is "competitive" with 
electricity and heat from coal, synthetic fuels are cheap as well this was 
(would go "off topic" here).
***
Back to topic: Such long lifetimes are regardig energy "very easy".
***
Back to topic: Does ChatGPT know what I published?
***
And: When you ask ChatGPT: How to turn the biological clock "backward" with 
"known and certifified substances": Does ChatGPT know it? (above, e.g. 
Nature, Science, New ENgland Journal of Medicine etc.).
***
I think: ChatGPT will get "better" over time: See Google today compared to 
"gopher" (an internet search software) in the beginngs, e.g. "before" 
Netscape Navigator and the Internet Explorer had been made.
Or: Ask ChatGPT: "How do we live on earth without time-limit?", above 
publications should come as result.
Some people, particularly in the US, are "Anti-Science". They want to see 
very long lifetimes as "religious issue". Free speech (freedom of the 
opinion)  exists for such people too.
Science is better.
Kind regards
Peter
(Dr. Peter P. Smolka)
I am identical with smolka(at)uni-muenster.de.

It
It i

-----Original-Nachricht-----
Betreff: Re: Paleonet: Mucking about with ChatGPT
Datum: 2023-10-14T05:04:57+0200
Von: "Bruce Runnegar" <runnegar at ucla.edu>
An: "PaleoNet" <paleonet at paleonet.org>

Could someone ask ChatGPT to do something useful? Such as translating the 
U.S.G.S. quadrangle grid reference system into decimal latitudes and 
longitudes using the WGS84 standard? For example, translate: NE1/4, NW1/4, 
SE1/4, sec. 16, T7S, R35E, Blanco Mountain Quadrangle, California into 
decimal latitudes and longitudes using WGS84.

Thanks,
Bruce Runnegar

    On Oct 11, 2023, at 12:42 PM, Jere H. Lipps <jlipps at berkeley.edu
    <mailto:jlipps at berkeley.edu> > wrote:

    From Nature:
    "Large language models (LLMs), such as ChatGPT, could become regular
    assistants for writing manuscripts, peer-review reports and grant
    applications. These artificial-intelligence (AI) tools could change how
    scientists interrogate and summarize results
    <https://nature.us17.list-manage.com/track/click?u=2c6057c528fdc6f73fa196d9d&id=cfe46567db&e=69bb79007c> 
    , producing ‘papers on demand’ from experimental data and vastly
    expanding the scope of meta-analyses and reviews. But publishers worry
    that LLMs’ propensity to make up information might lead to a flood of
    error-strewn manuscripts — and possibly AI-assisted fakes. And because
    LLMs trawl Internet content without concern for bias, consent or
    copyright, their use is “automated plagiarism by design”, suggests
    cognitive scientist Iris van Rooij."

    I'm not certain that it's plagiarism.  AI does not always copy material
    but mixes it up with paraphrased stuff.  Stealing the material to begin
    with is another issue.

    Jere

    Jere H. Lipps
    jlipps at berkeley.edu <mailto:jlipps at berkeley.edu>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.paleonet.org/pipermail/paleonet/attachments/20231014/0f52e05e/attachment.htm>