Taxacom: no artifical 'intelligence' yet
Tony Rees
tonyrees49 at gmail.com
Wed Feb 5 13:42:58 CST 2025
At the request of Dilrukshan Wijesinghe, I tried the same requests to
DeepSeek just now. For the genus *Monilinema* it was maybe 50% correct: it
new the genus name and placed it in the correct family in Cyanophyceae, but
got the year of description wrong (says 2013, actually 2020 or 2021) and
habitat wrong, also said "Monilinema species are often studied for their
ecological roles, such as their contributions to primary production and
nutrient cycling in aquatic ecosystems.", which is just plain wrong. On
Shirburn Castle it was much better than Chat GPT, giving a good overview
mostly in line with the Wikipedia account (my work), but heavily condensed
(which is fine). That account reads (if of interest):
----------------
*Shirburn Castle* is a historic fortified manor house located in *Shirburn*,
a village in Oxfordshire, England. It was built in the 14th century, around
1377, by *Warine de Lisle*, and it is a fine example of a medieval
quadrangular castle. The castle features a central courtyard surrounded by
four ranges of buildings, with towers at each corner, and it retains much
of its original medieval structure.
Shirburn Castle has a rich history and has been associated with several
notable families over the centuries. In the 18th century, it became the
seat of the *Earls of Macclesfield*, who made significant alterations to
the interior while preserving the castle's exterior appearance. The
Macclesfield family owned the castle for over 250 years and were known for
their contributions to science, particularly through the *Parker family*,
who were avid astronomers and collectors.
The castle is privately owned and not generally open to the public, but it
remains an important historical and architectural landmark. It is also
known for its extensive library, which was part of the Macclesfield
collection and included rare scientific manuscripts and books.
----------------
So no issues there.
Just a taster of DeepSeek vs Chat GPT at this time, 2 queries only of
course!
Regards - Tony
Tony Rees, New South Wales, Australia
https://nam10.safelinks.protection.outlook.com/?url=https%3A%2F%2Fabout.me%2FTonyRees&data=05%7C02%7Ctaxacom%40lists.ku.edu%7Ce0e3b5abd2fe4576967c08dd461d54b7%7C3c176536afe643f5b96636feabbe3c1a%7C0%7C0%7C638743815035632242%7CUnknown%7CTWFpbGZsb3d8eyJFbXB0eU1hcGkiOnRydWUsIlYiOiIwLjAuMDAwMCIsIlAiOiJXaW4zMiIsIkFOIjoiTWFpbCIsIldUIjoyfQ%3D%3D%7C40000%7C%7C%7C&sdata=RgpIsvRhMM9d0f3ByiIA9xu9YOZSQPRh5%2BeFFhBrRGo%3D&reserved=0
https://nam10.safelinks.protection.outlook.com/?url=http%3A%2F%2Fwww.irmng.org%2F&data=05%7C02%7Ctaxacom%40lists.ku.edu%7Ce0e3b5abd2fe4576967c08dd461d54b7%7C3c176536afe643f5b96636feabbe3c1a%7C0%7C0%7C638743815035653323%7CUnknown%7CTWFpbGZsb3d8eyJFbXB0eU1hcGkiOnRydWUsIlYiOiIwLjAuMDAwMCIsIlAiOiJXaW4zMiIsIkFOIjoiTWFpbCIsIldUIjoyfQ%3D%3D%7C40000%7C%7C%7C&sdata=oEToYPpuVR6Of5A1xh%2FE31W0saEFcwFOpRod9ex3QEQ%3D&reserved=0
On Thu, 6 Feb 2025 at 05:59, Tony Rees <tonyrees49 at gmail.com> wrote:
> As an aside from taxonomy, a few years back I did some "research"
> (information gathering and collation from published sources) on a historic
> building, Shirburn Castle in Oxfordshire, UK, which I used to update and
> considerably extend the previously existing Wikipedia entry on the same,
> see https://nam10.safelinks.protection.outlook.com/?url=https%3A%2F%2Fen.wikipedia.org%2Fwiki%2FShirburn_Castle&data=05%7C02%7Ctaxacom%40lists.ku.edu%7Ce0e3b5abd2fe4576967c08dd461d54b7%7C3c176536afe643f5b96636feabbe3c1a%7C0%7C0%7C638743815035666760%7CUnknown%7CTWFpbGZsb3d8eyJFbXB0eU1hcGkiOnRydWUsIlYiOiIwLjAuMDAwMCIsIlAiOiJXaW4zMiIsIkFOIjoiTWFpbCIsIldUIjoyfQ%3D%3D%7C40000%7C%7C%7C&sdata=m0cgv08Gou7%2FJmCxtmGY3jM1ONkGyOgJFk2UnTQ2iw4%3D&reserved=0 ... lots of good stuff
> there now, I would contend, all impeccably sourced. However Chat GPT simply
> says the following as at today:
> -----------
>
> Shirburn Castle is a historic country house located in Oxfordshire,
> England, near the village of Shirburn. It dates back to the 14th century
> and is one of the more well-preserved castles in the region. Over the
> centuries, it has served various purposes, including as a fortified
> residence and a seat of nobility.
>
> The castle is notable for its impressive architecture, which blends
> medieval and later periods, with much of the structure reflecting Tudor and
> Georgian styles. Though not as grand or famous as some other castles in
> England, Shirburn Castle has an interesting history and has been home to
> several influential families, most notably the Dunscombe family. In modern
> times, the estate is privately owned.
>
> One of the unique aspects of Shirburn Castle is its location within a
> beautiful and secluded area of the Oxfordshire countryside, making it a
> peaceful and picturesque spot. It's also been used as a venue for private
> events, weddings, and filming due to its atmospheric setting.
>
> Shirburn Castle is not typically open to the public, but it is an
> important part of the local history and offers an example of how English
> castles evolved over time.
> -----------
>
> Well, no mention of the very famous aspects of Shirburn at all, including
> its previously fabulous library and collections, seat of the Parker
> family/Earls of Macclesfield, etc. etc. - just one spurious reference to
> the Dunscombe family, no connection whatever. So not that impressive in
> this instance either.... just sayin' !!
>
>
> Regards to all - Tony
>
>
>
> On Thu, 6 Feb 2025 at 05:40, Tony Rees <tonyrees49 at gmail.com> wrote:
>
>> (Almost) every day I "research" notionally new (to me/my system)
>> taxonomic names (genera in the main) to see whether they really exist, and
>> are not some sort of lapsus or database error from elsewhere. This is an
>> area where AI could be very useful if it saves me trouble - normally I am
>> using Google Scholar, sector specific databases, occasionally regular
>> Google or Wikipedia to get other perspectives that I may have missed. Thus
>> Vlad's first point:
>>
>> 1. Taxonomic and Biodiversity Research
>> *Literature Mining & Data Extraction:*
>> AI can scan vast amounts of historical and current taxonomic literature
>> to extract species descriptions, synonymies, distribution data, and
>> morphological details. This can help build comprehensive databases and
>> identify trends or gaps in knowledge.
>>
>> I think for the better known names / taxa this is probably true, having
>> tried out a couple. However possibly not for less well known, or recently
>> established names. As an example (representative or not) I am up to
>> checking acritarch genera beginning with "M" in the "John Williams Index of
>> Palaeopalynology, placed online some time in the past 5 years or so I
>> believe, where I am up to "Monilinema Yin in Wang, 1984". You can see the
>> J. Williams record here: https://nam10.safelinks.protection.outlook.com/?url=https%3A%2F%2Fwww.mikrotax.org%2FAcritax%2FJWIP%2FMonilinema&data=05%7C02%7Ctaxacom%40lists.ku.edu%7Ce0e3b5abd2fe4576967c08dd461d54b7%7C3c176536afe643f5b96636feabbe3c1a%7C0%7C0%7C638743815035681610%7CUnknown%7CTWFpbGZsb3d8eyJFbXB0eU1hcGkiOnRydWUsIlYiOiIwLjAuMDAwMCIsIlAiOiJXaW4zMiIsIkFOIjoiTWFpbCIsIldUIjoyfQ%3D%3D%7C40000%7C%7C%7C&sdata=US4HrDV9jIyL7h%2Fp7%2F2S9PggNjBk7oXc4kJxyoBuZhM%3D&reserved=0
>> ... in this case Google knows about this, Google Scholar does not. However
>> Chat GPT does not either. In fact this is a nomen nudum according to
>> Williams. Now there is also a more recent genus Monilinema, a freshwater
>> cyanophyte (blue-green alga) from Brazil, described in 2020 in Journal of
>> Phycology, which is scarcely an obscure journal. This can be found easily
>> via both Google Scholar and regular Google; it is also in WoRMS (World
>> Register of Marine Species) as a freshwater taxon (data from AlgaeBase)
>> although not currently in Wikipedia. Again, Chat GPT is unaware of this.
>> Chat GPT did, however, tell me the following:
>> ------------------
>> The genus *Monilinema* refers to a group of parasitic nematodes in the
>> family *Mononchidae*, which are part of the order *Mononchida*. These
>> nematodes are generally predatory in nature, meaning they feed on other
>> nematodes and soil microorganisms rather than directly harming plants.
>> ------------------
>> which is completely incorrect (there is no genus *Monilinema* in
>> Nematoda; my own system tells me that the nearest matches there would be
>> Molinema Freitas & Lent, 1939 accepted as Dipetalonema Diesing, 1861, and
>> Monilonema Beveridge & Johnson, 1981, neither of which I have presently
>> classified in Mononchidae (but my classification might be out-of-date).
>>
>> So in terms of saving me work, not much use in this instance, although of
>> course there is always the option for Chat GPT to improve over time...
>>
>> Regards - Tony
>>
>> Tony Rees, New South Wales, Australia
>> https://nam10.safelinks.protection.outlook.com/?url=https%3A%2F%2Fabout.me%2FTonyRees&data=05%7C02%7Ctaxacom%40lists.ku.edu%7Ce0e3b5abd2fe4576967c08dd461d54b7%7C3c176536afe643f5b96636feabbe3c1a%7C0%7C0%7C638743815035696371%7CUnknown%7CTWFpbGZsb3d8eyJFbXB0eU1hcGkiOnRydWUsIlYiOiIwLjAuMDAwMCIsIlAiOiJXaW4zMiIsIkFOIjoiTWFpbCIsIldUIjoyfQ%3D%3D%7C40000%7C%7C%7C&sdata=Q4sJ1nkzvsnPkC4RaUmvSo5vV2Xb46EW%2FgSdMiy4w3E%3D&reserved=0
>> https://nam10.safelinks.protection.outlook.com/?url=http%3A%2F%2Fwww.irmng.org%2F&data=05%7C02%7Ctaxacom%40lists.ku.edu%7Ce0e3b5abd2fe4576967c08dd461d54b7%7C3c176536afe643f5b96636feabbe3c1a%7C0%7C0%7C638743815035712790%7CUnknown%7CTWFpbGZsb3d8eyJFbXB0eU1hcGkiOnRydWUsIlYiOiIwLjAuMDAwMCIsIlAiOiJXaW4zMiIsIkFOIjoiTWFpbCIsIldUIjoyfQ%3D%3D%7C40000%7C%7C%7C&sdata=ufCPmztgGpWyzAXqEekJ3bYfRqnNKaS7ijFeF3tBtR4%3D&reserved=0
>>
>>
>> On Wed, 5 Feb 2025 at 01:18, Vladimir Blagoderov via Taxacom <
>> taxacom at lists.ku.edu> wrote:
>>
>>> That's what ChatGPT has to say about the discussion :)
>>>
>>> Misconceptions About AI and LLMs in the Discussion
>>>
>>> 1.
>>>
>>> *AI as “Just a Metaphor” or “BS” Generator*
>>> - *What’s Said:* John Grehan opens by calling “artificial
>>> intelligence”
>>> a metaphor—implying it’s more a marketing construct than a real
>>> technology—and later labels its output as “BS.”
>>> - *The Reality:* While current LLMs are not sentient or
>>> “intelligent”
>>> in a human sense, they are based on robust statistical models
>>> that capture
>>> patterns in vast amounts of text. Their “intelligence” is not the
>>> same as
>>> human reasoning but can be extremely useful when its limitations
>>> are
>>> understood.
>>> 2.
>>>
>>> *Reliance Solely on Online Sources Makes AI Inherently Unreliable in
>>> Taxonomy*
>>> - *What’s Said:* Grehan argues that because many taxonomic articles
>>> aren’t open access or posted online, the AI’s sources are
>>> “limited” and
>>> thus its answers are suspect.
>>> - *The Reality:* Although LLMs are trained on large, publicly
>>> available datasets (and may miss some paywalled or less accessible
>>> literature), researchers are actively developing domain-specific
>>> models and
>>> fine-tuning methods that incorporate high-quality, specialized
>>> data.
>>> Moreover, even with general models, cross-checking and human
>>> expertise can
>>> compensate for gaps.
>>> 3.
>>>
>>> *Inconsistency Equals Incompetence*
>>> - *What’s Said:* Both John Grehan and Jared Bernard highlight
>>> inconsistent or contradictory responses (e.g., on whether
>>> *Magnificus*
>>> is monophyletic or the origin of panbiogeography) as evidence that
>>> AI “gets things wrong.”
>>> - *The Reality:* LLMs generate responses probabilistically based on
>>> the prompt and training data. Variations in phrasing or context
>>> can lead to
>>> different answers. Inconsistencies underscore the need for critical
>>> evaluation and context-aware querying rather than proving that the
>>> technology is entirely unreliable.
>>> 4.
>>>
>>> *AI as a Source of Plagiarism and a Threat to Independent Thought*
>>> - *What’s Said:* Jared Bernard expresses concern that LLMs
>>> “plagiarize”
>>> from papers and may foster a generation unable to think for itself.
>>> - *The Reality:* LLMs work by predicting text based on learned
>>> patterns—they sometimes echo phrases seen in training data without
>>> proper
>>> attribution. This is a recognized limitation rather than a
>>> deliberate act
>>> of plagiarism. Additionally, when used as an assistive tool rather
>>> than a
>>> crutch, AI can augment human creativity and efficiency rather
>>> than replace
>>> critical thinking.
>>> 5.
>>>
>>> *Expectation That AI Should Ask Clarifying Questions Like a Human
>>> Expert*
>>> - *What’s Said:* John Grehan criticizes the system for not asking
>>> whether he meant genus or species when he queried “Is Magnificus
>>> monophyletic?”
>>> - *The Reality:* Most current LLMs do not automatically ask
>>> clarifying questions unless specifically designed to do so. They
>>> instead
>>> generate the most statistically probable response given the prompt.
>>> Expecting interactive clarification mimics human
>>> conversation—but it isn’t
>>> a built‐in feature of many LLMs as they’re typically deployed
>>> today.
>>> 6.
>>>
>>> *Belief That Advancements Like “Deep Research” Will Replace
>>> Researchers
>>> Entirely*
>>> - *What’s Said:* Jared Bernard mentions that OpenAI’s “Deep Research”
>>> might eventually replace researchers.
>>> - *The Reality:* While automation in literature synthesis and data
>>> analysis is progressing, the goal is to assist and enhance human
>>> research—not to eliminate the need for critical human judgment and
>>> domain
>>> expertise.
>>>
>>> ------------------------------
>>> Where John Grehan Is Wrong
>>>
>>> 1.
>>>
>>> *Misunderstanding the Nature of LLM Responses:*
>>> - *Issue:* Grehan’s criticism that the answer to his query was “BS”
>>> because it did not clarify the taxonomic level (genus vs. species)
>>> overlooks the design of LLMs.
>>> - *Clarification:* LLMs work by predicting likely continuations of
>>> text. They do not “know” to ask clarifying questions unless they’re
>>> specifically designed or prompted to do so. Expecting an LLM to
>>> mirror the
>>> interactive, context-sensitive process of an expert taxonomist
>>> sets an
>>> unrealistically high bar for current technology.
>>> 2.
>>>
>>> *Overgeneralizing from Inconsistent Responses:*
>>> - *Issue:* Grehan cites different responses (one stating *Magnificus*
>>> is
>>> polyphyletic, another monophyletic) as proof of AI’s inadequacy.
>>> - *Clarification:* Variability in output can occur due to
>>> differences
>>> in the underlying model version, prompt phrasing, or the
>>> probabilistic
>>> nature of text generation. Rather than showing that the
>>> technology “doesn’t
>>> work,” such inconsistencies highlight the importance of using AI
>>> outputs as
>>> starting points that require expert validation and refinement.
>>> 3.
>>>
>>> *Assuming Limited Data Equals Inherent Unreliability:*
>>> - *Issue:* Grehan’s assertion that because AI may not access certain
>>> paywalled or offline articles, its answers are “BS.”
>>> - *Clarification:* While access to comprehensive data is crucial,
>>> ongoing efforts in research are aimed at integrating more
>>> specialized
>>> datasets into models. Moreover, even imperfect outputs can be
>>> valuable when
>>> used in combination with expert oversight.
>>>
>>> ------------------------------
>>> How AI Can Be Used in Taxonomic and Biodiversity Research
>>>
>>> Despite the concerns raised, AI (and LLMs in particular) holds
>>> considerable
>>> promise for enhancing work in taxonomy, species description, and related
>>> fields. Here are several key applications:
>>> 1. Taxonomic and Biodiversity Research
>>>
>>> -
>>>
>>> *Literature Mining & Data Extraction:*
>>> AI can scan vast amounts of historical and current taxonomic
>>> literature
>>> to extract species descriptions, synonymies, distribution data, and
>>> morphological details. This can help build comprehensive databases and
>>> identify trends or gaps in knowledge.
>>> -
>>>
>>> *Data Integration:*
>>> Machine learning algorithms can integrate data from diverse sources
>>> (e.g., genetic sequences, morphological records, ecological data) to
>>> provide a more holistic view of biodiversity patterns.
>>> -
>>>
>>> *Species Distribution Modeling:*
>>> AI-driven geospatial analysis can help predict species distributions
>>> and
>>> assess changes over time, which is crucial for conservation planning
>>> and
>>> understanding the effects of climate change.
>>>
>>> 2. New Species Descriptions
>>>
>>> -
>>>
>>> *Drafting Descriptions:*
>>> LLMs can generate preliminary drafts of species descriptions by
>>> synthesizing available data. Taxonomists can then refine these drafts,
>>> ensuring adherence to standard nomenclatural protocols.
>>> -
>>>
>>> *Cross-Referencing:*
>>> Automated tools can compare new findings against existing literature
>>> to
>>> help confirm the novelty of a species and avoid redundant
>>> descriptions.
>>>
>>> 3. Taxonomic Monographs
>>>
>>> -
>>>
>>> *Compiling and Organizing Data:*
>>> AI systems can assist in compiling literature, specimen data, and
>>> previous research findings into structured formats, saving time in the
>>> preparation of comprehensive monographs.
>>> -
>>>
>>> *Automated Consistency Checks:*
>>> Tools can be designed to check for nomenclatural consistency, verify
>>> synonymies, and flag discrepancies within large taxonomic
>>> compilations.
>>>
>>> 4. Checklists
>>>
>>> -
>>>
>>> *Dynamic Data Aggregation:*
>>> AI can continuously harvest and integrate data from museums,
>>> biodiversity databases (like GBIF), and published literature to keep
>>> species checklists up to date.
>>> -
>>>
>>> *Error Detection & Standardization:*
>>> Machine learning algorithms can identify potential errors or outdated
>>> classifications in checklists, prompting updates and ensuring
>>> consistency
>>> across databases.
>>>
>>> 5. Phylogenetic Research
>>>
>>> -
>>>
>>> *Data Analysis and Tree Inference:*
>>> AI algorithms can process large-scale genetic and morphological
>>> datasets
>>> to assist in constructing phylogenetic trees. They can help manage the
>>> computational complexity involved in large datasets.
>>> -
>>>
>>> *Integrating Multiple Data Types:*
>>> By combining molecular data with morphological and ecological data, AI
>>> can help reconcile discrepancies and suggest more robust phylogenetic
>>> hypotheses.
>>> -
>>>
>>> *Predictive Modeling:*
>>> Machine learning models can be used to predict evolutionary
>>> relationships and diversification patterns, offering new hypotheses
>>> for
>>> empirical testing.
>>>
>>> ------------------------------
>>> Conclusion
>>>
>>> The concerns expressed in the email chain largely stem from
>>> misconceptions
>>> about how AI and LLMs work and what their current limitations are. While
>>> it’s true that LLMs do not “think” or interact in the nuanced way that
>>> human experts do, they offer substantial potential to streamline and
>>> enhance various aspects of taxonomic and biodiversity research. Rather
>>> than
>>> replacing experts, AI tools can serve as valuable assistants—automating
>>> routine tasks, synthesizing large datasets, and even helping draft
>>> scientific texts—while leaving the critical interpretation and
>>> decision-making to trained researchers. As the technology continues to
>>> improve and more domain-specific models are developed, its contributions
>>> to
>>> fields like taxonomy are likely to grow even more significant.
>>>
>>>
>>> Cheers,
>>>
>>> Vlad
>>>
>>> --
>>> Dr Vladimir Blagoderov, FLS, FRMS
>>> Principal curator – Invertebrates
>>>
>>> National Museums Scotland
>>>
>>> Chambers Street
>>>
>>> Edinburgh
>>>
>>> EH1 1JF
>>>
>>> National Museums Collection Centre
>>>
>>> 242 West Granton Road
>>>
>>> Edinburgh
>>>
>>> EH5 1JA
>>>
>>>
>>>
>>> tel. +44(0)131 247 4261
>>>
>>>
>>> e-mail:
>>>
>>> *v.blagoderov at nms.ac.uk <vblagoderov at nms.ac.uk>*
>>> vblago at gmail.com
>>>
>>> Fungus Gnats Online:
>>> https://nam10.safelinks.protection.outlook.com/?url=http%3A%2F%2Fwww.sciaroidea.info%2F&data=05%7C02%7Ctaxacom%40lists.ku.edu%7Ce0e3b5abd2fe4576967c08dd461d54b7%7C3c176536afe643f5b96636feabbe3c1a%7C0%7C0%7C638743815035726305%7CUnknown%7CTWFpbGZsb3d8eyJFbXB0eU1hcGkiOnRydWUsIlYiOiIwLjAuMDAwMCIsIlAiOiJXaW4zMiIsIkFOIjoiTWFpbCIsIldUIjoyfQ%3D%3D%7C40000%7C%7C%7C&sdata=wKmt2z0sAFiX9n984bixls80gv0miGOg04S7WxM5Yw4%3D&reserved=0
>>>
>>> On Mon, 3 Feb 2025 at 21:28, Dilrukshan Wijesinghe via Taxacom <
>>> taxacom at lists.ku.edu> wrote:
>>>
>>> > Would be interesting to see the results from posing the same real
>>> > taxonomic questions to the most recent incarnation of OpenAI’s product
>>> and
>>> > DeepSeek.
>>> >
>>> > Priyantha
>>> >
>>> >
>>> > Sent from my iPhone
>>> >
>>> > > On Feb 3, 2025, at 3:41 PM, John Grehan via Taxacom <
>>> > taxacom at lists.ku.edu> wrote:
>>> > >
>>> > > Actually my first question was 'Is Magnificus monophyletic' The
>>> response
>>> > > only referenced a species, not the genus (which is what I had in
>>> mind).
>>> > > Something truly intelligent (and acquainted with taxonomy) would have
>>> > asked
>>> > > what taxonomic level, or at least whether genus or species, or even
>>> gave
>>> > > both (which ironically is likely to happen if one typed Magnificus
>>> as a
>>> > > keyword into a standard Google or equivalent search). I guess
>>> > > eventually computer programming is going to get sophisticated enough
>>> to
>>> > > write entire taxonomic papers - just give the data, or eventually a
>>> > robotic
>>> > > computer will also handle all the specimens and dissect, map DNA etc
>>> > > without anyone being involved at all. Then there will be no human
>>> > > taxonomists left to read the result. Of course if we're lucky some
>>> nut
>>> > will
>>> > > blow the world up before then.
>>> > >
>>> > > Cheers, John
>>> > >
>>> > >> On Mon, Feb 3, 2025 at 3:32 PM Scott Thomson <
>>> > scott.thomson321 at gmail.com>
>>> > >> wrote:
>>> > >>
>>> > >> Putting on my programmer hat for sec......
>>> > >>
>>> > >> The AI's are as good as they are taught to be. It takes time to
>>> "teach"
>>> > >> them the information. The information being loaded into them is
>>> also not
>>> > >> even. Maths and Engineering are first our fields not so much. I have
>>> > >> dabbled every now and then writing prompts to teach ChatGTP
>>> taxonomic
>>> > >> information. Remember when asking an AI for information you have to
>>> run
>>> > a
>>> > >> series of prompts to effectively teach it fields it has less
>>> > information on.
>>> > >>
>>> > >> Cheers Scott
>>> > >>
>>> > >> On Mon, Feb 3, 2025 at 5:25 PM John Grehan via Taxacom <
>>> > >> taxacom at lists.ku.edu> wrote:
>>> > >>
>>> > >>> Just goes to show that whatever one might use one might get
>>> different
>>> > >>> results.Makes my point.
>>> > >>>
>>> > >>> On Mon, Feb 3, 2025 at 3:19 PM Roderic Page via Taxacom <
>>> > >>> taxacom at lists.ku.edu> wrote:
>>> > >>>
>>> > >>>> Hi John,
>>> > >>>>
>>> > >>>> You might want to try that again. Using ChatGPT 4o I get:
>>> > >>>>
>>> > >>>> "Yes, the genus Magnificus is considered monophyletic. A 2021
>>> study by
>>> > >>>> Grehan et al. provided evidence supporting this by identifying
>>> unique
>>> > >>>> morphological features shared among species within the genus.
>>> > >>> Specifically,
>>> > >>>> they noted a distinctive shape of the ergosterol sclerite in the
>>> > >>> anterior
>>> > >>>> abdominal region and a unique pseudoteguminal arm in the male
>>> > genitalia.
>>> > >>>> These shared characteristics suggest that all species within the
>>> genus
>>> > >>>> Magnificus descend from a common ancestor, supporting its
>>> monophyly.”
>>> > >>>>
>>> > >>>> The answer includes a link to your paper in PubMed
>>> > >>>> https://nam10.safelinks.protection.outlook.com/?url=https%3A%2F%2Fpubmed.ncbi.nlm.nih.gov%2F33756654%2F&data=05%7C02%7Ctaxacom%40lists.ku.edu%7Ce0e3b5abd2fe4576967c08dd461d54b7%7C3c176536afe643f5b96636feabbe3c1a%7C0%7C0%7C638743815035743426%7CUnknown%7CTWFpbGZsb3d8eyJFbXB0eU1hcGkiOnRydWUsIlYiOiIwLjAuMDAwMCIsIlAiOiJXaW4zMiIsIkFOIjoiTWFpbCIsIldUIjoyfQ%3D%3D%7C40000%7C%7C%7C&sdata=3k6iSwi5ib07PkulVJ%2FXoRSHiBEjhKv7Sk0DdcgZBRo%3D&reserved=0
>>> > >>>>
>>> > >>>> Things are moving so fast that I'd resist making sweeping
>>> > >>> generalisations
>>> > >>>> just yet...
>>> > >>>>
>>> > >>>> Regards,
>>> > >>>>
>>> > >>>> Rod
>>> > >>>>
>>> > >>>> ---------------------------------------------------------
>>> > >>>> Roderic Page
>>> > >>>> Professor of Taxonomy
>>> > >>>> School of Biodiversity, One Health, & Veterinary Medicine
>>> > >>>> College of Medical, Veterinary and Life Sciences
>>> > >>>> Graham Kerr Building
>>> > >>>> University of Glasgow
>>> > >>>> Glasgow G12 8QQ, UK
>>> > >>>>
>>> > >>>> Email: Roderic.Page at glasgow.ac.uk<mailto:
>>> Roderic.Page at glasgow.ac.uk>
>>> > >>>> LinkedIn: https://nam10.safelinks.protection.outlook.com/?url=https%3A%2F%2Fuk.linkedin.com%2Fin%2Frdmpage&data=05%7C02%7Ctaxacom%40lists.ku.edu%7Ce0e3b5abd2fe4576967c08dd461d54b7%7C3c176536afe643f5b96636feabbe3c1a%7C0%7C0%7C638743815035761303%7CUnknown%7CTWFpbGZsb3d8eyJFbXB0eU1hcGkiOnRydWUsIlYiOiIwLjAuMDAwMCIsIlAiOiJXaW4zMiIsIkFOIjoiTWFpbCIsIldUIjoyfQ%3D%3D%7C40000%7C%7C%7C&sdata=v3eW%2F1vLjBZ4Gwj%2FWDogtsKShN6y2GFxXHAa8ou5eRM%3D&reserved=0
>>> > >>>> Twitter: https://nam10.safelinks.protection.outlook.com/?url=https%3A%2F%2Ftwitter.com%2Frdmpage&data=05%7C02%7Ctaxacom%40lists.ku.edu%7Ce0e3b5abd2fe4576967c08dd461d54b7%7C3c176536afe643f5b96636feabbe3c1a%7C0%7C0%7C638743815035778990%7CUnknown%7CTWFpbGZsb3d8eyJFbXB0eU1hcGkiOnRydWUsIlYiOiIwLjAuMDAwMCIsIlAiOiJXaW4zMiIsIkFOIjoiTWFpbCIsIldUIjoyfQ%3D%3D%7C40000%7C%7C%7C&sdata=XHxPo99tdqkzYTpcnw%2BEzRWl4bAmSpdxsmMuL3DF8%2F8%3D&reserved=0
>>> > >>>> Telegram: rdmpage
>>> > >>>> Blog: https://nam10.safelinks.protection.outlook.com/?url=https%3A%2F%2Fiphylo.blogspot.com%2F&data=05%7C02%7Ctaxacom%40lists.ku.edu%7Ce0e3b5abd2fe4576967c08dd461d54b7%7C3c176536afe643f5b96636feabbe3c1a%7C0%7C0%7C638743815035796924%7CUnknown%7CTWFpbGZsb3d8eyJFbXB0eU1hcGkiOnRydWUsIlYiOiIwLjAuMDAwMCIsIlAiOiJXaW4zMiIsIkFOIjoiTWFpbCIsIldUIjoyfQ%3D%3D%7C40000%7C%7C%7C&sdata=v65yEplLMhnUX4XLYnm16y2Kf%2BQNz9h6jpX9CVIyWOY%3D&reserved=0<https://nam10.safelinks.protection.outlook.com/?url=https%3A%2F%2Fiphylo.blogspot.com%2F&data=05%7C02%7Ctaxacom%40lists.ku.edu%7Ce0e3b5abd2fe4576967c08dd461d54b7%7C3c176536afe643f5b96636feabbe3c1a%7C0%7C0%7C638743815035814922%7CUnknown%7CTWFpbGZsb3d8eyJFbXB0eU1hcGkiOnRydWUsIlYiOiIwLjAuMDAwMCIsIlAiOiJXaW4zMiIsIkFOIjoiTWFpbCIsIldUIjoyfQ%3D%3D%7C40000%7C%7C%7C&sdata=vDmrwskUwz8pWVNnRUEzobeemQGX39m87%2FjaO10Zybc%3D&reserved=0>
>>> > >>>> ORCID: https://nam10.safelinks.protection.outlook.com/?url=https%3A%2F%2Forcid.org%2F0000-0002-7101-9767&data=05%7C02%7Ctaxacom%40lists.ku.edu%7Ce0e3b5abd2fe4576967c08dd461d54b7%7C3c176536afe643f5b96636feabbe3c1a%7C0%7C0%7C638743815035831971%7CUnknown%7CTWFpbGZsb3d8eyJFbXB0eU1hcGkiOnRydWUsIlYiOiIwLjAuMDAwMCIsIlAiOiJXaW4zMiIsIkFOIjoiTWFpbCIsIldUIjoyfQ%3D%3D%7C40000%7C%7C%7C&sdata=UXzn3Fd6ItZHaN%2FUB%2F%2BJqoaf1fKuU95gHJfVowmSe2A%3D&reserved=0
>>> > >>>> Citations:
>>> > >>> https://nam10.safelinks.protection.outlook.com/?url=https%3A%2F%2Fscholar.google.co.uk%2Fcitations%3Fhl%3Den%26user%3D4Z5WABAAAAAJ&data=05%7C02%7Ctaxacom%40lists.ku.edu%7Ce0e3b5abd2fe4576967c08dd461d54b7%7C3c176536afe643f5b96636feabbe3c1a%7C0%7C0%7C638743815035846782%7CUnknown%7CTWFpbGZsb3d8eyJFbXB0eU1hcGkiOnRydWUsIlYiOiIwLjAuMDAwMCIsIlAiOiJXaW4zMiIsIkFOIjoiTWFpbCIsIldUIjoyfQ%3D%3D%7C40000%7C%7C%7C&sdata=FZXRkPBbU%2BZnbGDJQjkHyKaBXbFuBeFw%2F2rxkW%2B%2FLlI%3D&reserved=0
>>> > >>>> ResearchGate https://nam10.safelinks.protection.outlook.com/?url=https%3A%2F%2Fwww.researchgate.net%2Fprofile%2FRoderic_Page&data=05%7C02%7Ctaxacom%40lists.ku.edu%7Ce0e3b5abd2fe4576967c08dd461d54b7%7C3c176536afe643f5b96636feabbe3c1a%7C0%7C0%7C638743815035859628%7CUnknown%7CTWFpbGZsb3d8eyJFbXB0eU1hcGkiOnRydWUsIlYiOiIwLjAuMDAwMCIsIlAiOiJXaW4zMiIsIkFOIjoiTWFpbCIsIldUIjoyfQ%3D%3D%7C40000%7C%7C%7C&sdata=%2FV9pp%2BXPa%2F%2FJwt9pI%2F53amaLdTYZo8Atnk9nVANQB8E%3D&reserved=0
>>> > >>>> On 3 Feb 2025 at 19:46 +0000, John Grehan via Taxacom <
>>> > >>>> taxacom at lists.ku.edu>, wrote:
>>> > >>>> I find 'artificial intelligence' to be one of those metaphors
>>> that is
>>> > >>> more
>>> > >>>> about creating a state of reality than reality itself (a bit like
>>> the
>>> > >>>> 'barcode' metaphor).
>>> > >>>>
>>> > >>>> Just for the heck of it, I asked Chatgpg "is the genus Magnificus
>>> > >>>> monophyletic?" (this is a genus of ghost moths). Answer: "Based on
>>> > >>> current
>>> > >>>> research and taxonomic classification, *Magnificus* is *not
>>> considered
>>> > >>>> monophyletic*. In fact, it’s generally thought to be polyphyletic,
>>> > >>> meaning
>>> > >>>> that the species within the genus do not all share a single common
>>> > >>> ancestor
>>> > >>>> exclusive to that group. They are often found to be more closely
>>> > >>> related to
>>> > >>>> species in other genera than to each other."
>>> > >>>>
>>> > >>>> This is about as much BS that anyone could invent. I think it
>>> shows
>>> > that
>>> > >>>> the 'intelligence' of a computer is only as good as its sources. I
>>> > >>> expect
>>> > >>>> in taxonomy its sources are quite limited - especially if
>>> articles are
>>> > >>> not
>>> > >>>> open access or otherwise posted on the web.
>>> > >>>>
>>> > >>>> In another time wasting exercises (relatively speaking I don't
>>> have
>>> > that
>>> > >>>> much time left, so might as well waste some of it :) I asked
>>> 'what is
>>> > >>>> panbiogeography' . In one answer I was told that I created
>>> > >>> panbiogeography,
>>> > >>>> in another question it was Michael Heads who created it. Of course
>>> > >>> neither
>>> > >>>> is correct, and I guess 1958 sinks to the bottom of the swamp
>>> where AI
>>> > >>> does
>>> > >>>> not dare to tread.
>>> > >>>>
>>> > >>>> Is there no hope for civilization?
>>> > >>>>
>>> > >>>> Cheers,
>>> > >>>>
>>> > >>>> John Grehan
>>> > >>>>
>>> > >>>> --
>>> > >>>> https://nam10.safelinks.protection.outlook.com/?url=https%3A%2F%2Fhepialidsoftheworld.com.au%2F&data=05%7C02%7Ctaxacom%40lists.ku.edu%7Ce0e3b5abd2fe4576967c08dd461d54b7%7C3c176536afe643f5b96636feabbe3c1a%7C0%7C0%7C638743815035873024%7CUnknown%7CTWFpbGZsb3d8eyJFbXB0eU1hcGkiOnRydWUsIlYiOiIwLjAuMDAwMCIsIlAiOiJXaW4zMiIsIkFOIjoiTWFpbCIsIldUIjoyfQ%3D%3D%7C40000%7C%7C%7C&sdata=oac8QOg4gORSgl4cSLWhBDGqa8SBc86gGGK98Xr2Lzc%3D&reserved=0 (use the 'visit archived web
>>> > site'
>>> > >>>> link, then the 'Ghost Moth Research page' link.
>>> > >>>> _______________________________________________
>>> > >>>> Taxacom Mailing List
>>> > >>>>
>>> > >>>> Nurturing nuance while assailing ambiguity for 38 years,
>>> 1987-2025.
>>> > >>>>
>>> > >>>> Send Taxacom mailing list submissions to: taxacom at lists.ku.edu
>>> > >>>> For list information; to subscribe or unsubscribe, visit:
>>> > >>>> https://lists.ku.edu/listinfo/taxacom
>>> > >>>> You can reach the person managing the list at:
>>> > >>> taxacom-owner at lists.ku.edu
>>> > >>>>
>>> > >>>>
>>> > >>>> _______________________________________________
>>> > >>>> Taxacom Mailing List
>>> > >>>>
>>> > >>>> Nurturing nuance while assailing ambiguity for 38 years,
>>> 1987-2025.
>>> > >>>>
>>> > >>>> Send Taxacom mailing list submissions to: taxacom at lists.ku.edu
>>> > >>>> For list information; to subscribe or unsubscribe, visit:
>>> > >>>> https://lists.ku.edu/listinfo/taxacom
>>> > >>>> You can reach the person managing the list at:
>>> > >>> taxacom-owner at lists.ku.edu
>>> > >>>>
>>> > >>>>
>>> > >>>>
>>> > >>>
>>> > >>> --
>>> > >>> https://nam10.safelinks.protection.outlook.com/?url=https%3A%2F%2Fhepialidsoftheworld.com.au%2F&data=05%7C02%7Ctaxacom%40lists.ku.edu%7Ce0e3b5abd2fe4576967c08dd461d54b7%7C3c176536afe643f5b96636feabbe3c1a%7C0%7C0%7C638743815035885545%7CUnknown%7CTWFpbGZsb3d8eyJFbXB0eU1hcGkiOnRydWUsIlYiOiIwLjAuMDAwMCIsIlAiOiJXaW4zMiIsIkFOIjoiTWFpbCIsIldUIjoyfQ%3D%3D%7C40000%7C%7C%7C&sdata=ut1PZNyiQtWOOQfR7JfHJyMdTM71NB7%2F22lbaJY2OWk%3D&reserved=0 (use the 'visit archived web
>>> site'
>>> > >>> link, then the 'Ghost Moth Research page' link.
>>> > >>> _______________________________________________
>>> > >>> Taxacom Mailing List
>>> > >>>
>>> > >>> Nurturing nuance while assailing ambiguity for 38 years, 1987-2025.
>>> > >>>
>>> > >>> Send Taxacom mailing list submissions to: taxacom at lists.ku.edu
>>> > >>> For list information; to subscribe or unsubscribe, visit:
>>> > >>> https://lists.ku.edu/listinfo/taxacom
>>> > >>> You can reach the person managing the list at:
>>> > taxacom-owner at lists.ku.edu
>>> > >>>
>>> > >>>
>>> > >>>
>>> > >>
>>> > >> --
>>> > >> Scott Thomson
>>> > >>
>>> > >> Centro de Estudos dos Quelônios da Amazônia - CEQUA
>>> > >> Petrópolis, Manaus
>>> > >> State of Amazonas, 69055-010
>>> > >> Brasil
>>> > >>
>>> > >> https://nam10.safelinks.protection.outlook.com/?url=http%3A%2F%2Fwww.carettochelys.com%2F&data=05%7C02%7Ctaxacom%40lists.ku.edu%7Ce0e3b5abd2fe4576967c08dd461d54b7%7C3c176536afe643f5b96636feabbe3c1a%7C0%7C0%7C638743815035900336%7CUnknown%7CTWFpbGZsb3d8eyJFbXB0eU1hcGkiOnRydWUsIlYiOiIwLjAuMDAwMCIsIlAiOiJXaW4zMiIsIkFOIjoiTWFpbCIsIldUIjoyfQ%3D%3D%7C40000%7C%7C%7C&sdata=MaHnYzoQjNBJK8BwvPLS2aSR%2FJm%2FJRs5yzqXBI6oRXI%3D&reserved=0
>>> > >> ORCID: https://nam10.safelinks.protection.outlook.com/?url=http%3A%2F%2Forcid.org%2F0000-0003-1279-2722&data=05%7C02%7Ctaxacom%40lists.ku.edu%7Ce0e3b5abd2fe4576967c08dd461d54b7%7C3c176536afe643f5b96636feabbe3c1a%7C0%7C0%7C638743815035912349%7CUnknown%7CTWFpbGZsb3d8eyJFbXB0eU1hcGkiOnRydWUsIlYiOiIwLjAuMDAwMCIsIlAiOiJXaW4zMiIsIkFOIjoiTWFpbCIsIldUIjoyfQ%3D%3D%7C40000%7C%7C%7C&sdata=cv1tmjh7TzFTWmpnQ3yMGA%2BtZzNkri58xp4Mhyw7QRQ%3D&reserved=0
>>> > >> Lattes: *https://nam10.safelinks.protection.outlook.com/?url=http%3A%2F%2Flattes.cnpq.br%2F0323517916624728*&data=05%7C02%7Ctaxacom%40lists.ku.edu%7Ce0e3b5abd2fe4576967c08dd461d54b7%7C3c176536afe643f5b96636feabbe3c1a%7C0%7C0%7C638743815035924347%7CUnknown%7CTWFpbGZsb3d8eyJFbXB0eU1hcGkiOnRydWUsIlYiOiIwLjAuMDAwMCIsIlAiOiJXaW4zMiIsIkFOIjoiTWFpbCIsIldUIjoyfQ%3D%3D%7C40000%7C%7C%7C&sdata=u9ZSx7jsWMe9M98ydEVprJ2nu3GlvywIvExxZXhc8Fc%3D&reserved=0
>>> > >> <
>>> >
>>> https://nam10.safelinks.protection.outlook.com/?url=https%3A%2F%2Fwwws.cnpq.br%2Fcvlattesweb%2FPKG_MENU.menu%3Ff_cod%3D1E409F4BF37BFC4AD13FD58CDB7AA5FD%23&data=05%7C02%7Ctaxacom%40lists.ku.edu%7Ce0e3b5abd2fe4576967c08dd461d54b7%7C3c176536afe643f5b96636feabbe3c1a%7C0%7C0%7C638743815035937714%7CUnknown%7CTWFpbGZsb3d8eyJFbXB0eU1hcGkiOnRydWUsIlYiOiIwLjAuMDAwMCIsIlAiOiJXaW4zMiIsIkFOIjoiTWFpbCIsIldUIjoyfQ%3D%3D%7C40000%7C%7C%7C&sdata=WR4tB1IzgYQrynRekVaQ3HnBRkm2WYb0MU2h5VxRhvU%3D&reserved=0
>>> > >
>>> > >> Skype: Faendalimas
>>> > >> Mobile Phone Brasil: +55 11 98178 7270
>>> > >> Whatsapp: +55 11 98178 7270
>>> > >>
>>> > >>
>>> > >
>>> > > --
>>> > > https://nam10.safelinks.protection.outlook.com/?url=https%3A%2F%2Fhepialidsoftheworld.com.au%2F&data=05%7C02%7Ctaxacom%40lists.ku.edu%7Ce0e3b5abd2fe4576967c08dd461d54b7%7C3c176536afe643f5b96636feabbe3c1a%7C0%7C0%7C638743815035952254%7CUnknown%7CTWFpbGZsb3d8eyJFbXB0eU1hcGkiOnRydWUsIlYiOiIwLjAuMDAwMCIsIlAiOiJXaW4zMiIsIkFOIjoiTWFpbCIsIldUIjoyfQ%3D%3D%7C40000%7C%7C%7C&sdata=zXj1hjgl3WgO3VB7NMPveLORc%2Bh%2BZM3dfy41HPMee%2Bg%3D&reserved=0 (use the 'visit archived web
>>> site'
>>> > > link, then the 'Ghost Moth Research page' link.
>>> > > _______________________________________________
>>> > > Taxacom Mailing List
>>> > >
>>> > > Nurturing nuance while assailing ambiguity for 38 years, 1987-2025.
>>> > >
>>> > > Send Taxacom mailing list submissions to: taxacom at lists.ku.edu
>>> > > For list information; to subscribe or unsubscribe, visit:
>>> > https://lists.ku.edu/listinfo/taxacom
>>> > > You can reach the person managing the list at:
>>> > taxacom-owner at lists.ku.edu
>>> > >
>>> > >
>>> >
>>> > _______________________________________________
>>> > Taxacom Mailing List
>>> >
>>> > Nurturing nuance while assailing ambiguity for 38 years, 1987-2025.
>>> >
>>> > Send Taxacom mailing list submissions to: taxacom at lists.ku.edu
>>> > For list information; to subscribe or unsubscribe, visit:
>>> > https://lists.ku.edu/listinfo/taxacom
>>> > You can reach the person managing the list at:
>>> taxacom-owner at lists.ku.edu
>>> >
>>> >
>>> >
>>> _______________________________________________
>>> Taxacom Mailing List
>>>
>>> Nurturing nuance while assailing ambiguity for 38 years, 1987-2025.
>>>
>>> Send Taxacom mailing list submissions to: taxacom at lists.ku.edu
>>> For list information; to subscribe or unsubscribe, visit:
>>> https://lists.ku.edu/listinfo/taxacom
>>> You can reach the person managing the list at:
>>> taxacom-owner at lists.ku.edu
>>>
>>>
>>>
More information about the Taxacom
mailing list