T&D - AI

45 views
Skip to first unread message

dwa...@advaita.org.uk

unread,
Jan 12, 2026, 3:37:31 AM (7 days ago) Jan 12
to adva...@googlegroups.com

Regarding the comments on the value of AI:

 

I do agree that AI can be excellent in clarifying points in Advaita BUT…

As I (and others) have pointed out before, one needs to be very careful about how much to believe.

 

I have just begun posting a T&D for anvaya-vyatireka to the website. Initially, I was going to omit this because I thought it would only be 2 or 3 paragraphs. But a chance encounter with a Śaṅkara quotation from the Gaudapada kArikA bhAShya caused me to ask ChatGPT about the interpretation, since I did not think that Alston’s translation could be strictly accurate. That discussion has increased the size of the post to (currently) over 8000 words. And – the point I wish to make here – is that the AI, despite having made two attempts to explain GK3.31 – and doing so brilliantly – was still ‘making it all up’. Even after I pointed it out the first time, it still made something else up the second time! In the end, I had to provide the Sanskrit for it to use. It aways apologizes profusely but, needless to say, that does not help.

 

Anyway, I will not post this definition here. Those who want to follow the ‘discussion’ can see it in full at the website (but note that I am only posting around 1500 words per week, so that it will take a couple of months before the complete discussion is available. This note is really in the way of another warning about being very careful in what you believe from AI.

 

Best wishes,

Dennis

 

Sudhanshu Shekhar

unread,
Jan 12, 2026, 7:10:53 AM (7 days ago) Jan 12
to adva...@googlegroups.com
Namaste Dennis prabhu ji.

 I have just begun posting a T&D for anvaya-vyatireka to the website. Initially, I was going to omit this because I thought it would only be 2 or 3 paragraphs. But a chance encounter with a Śaṅkara quotation from the Gaudapada kArikA bhAShya caused me to ask ChatGPT about the interpretation, since I did not think that Alston’s translation could be strictly accurate. That discussion has increased the size of the post to (currently) over 8000 words. And – the point I wish to make here – is that the AI, despite having made two attempts to explain GK3.31 – and doing so brilliantly – was still ‘making it all up’. Even after I pointed it out the first time, it still made something else up the second time! In the end, I had to provide the Sanskrit for it to use. It aways apologizes profusely but, needless to say, that does not help.


Actually, anvaya-vyatireka of advaita and that of NyAya are a bit different. 

In advaita, anvaya-vyatireka is used to prove the causality of A with respect to B. 

So, the language used is:

1. if A is present, B is present. (anvaya) 
2. if A is absent, B is absent. (vyatireka).

If this is so, then A is cause of B. This is held in advaita. A classical example of this is - मनोदृश्यम् इदं द्वैतं सर्वं मन इति प्रतिज्ञा, तद्भावे भावात् तदभावे चाभावात्

When I had read it first, I was utterly confused. This just cannot be, I thought. For e.g. if clay is, it cannot be guaranteed that pot is also present. Clay can be in the form of lump also. So, presence of cause is no guarantee for presence of pot. 

NyAya, however, would go in another direction. They will say, if A is cause of B, then:

1. if B is present, then A is present. [कार्यस्य भावे, कारणस्य भावः]
2. if A is absent, then B is absent. [कारणस्य अभावे, कार्यस्य अभावः]

This is their version of anvaya-vyatireka. 

So, I was just unable to understand BhAshyakAra's statement. I hope I am able to communicate the confusion.

The confusion is cleared elsewhere (I will share the reference). Actually, a word "एव" needs to be introduced to understand advaitic anvaya-vyatireka.

So, Advaita would say: 

1. कारणस्य भावे (एव), कार्यस्य भावः [anvaya] - [(Only when) kAraNa is present, kArya is present.]
2. कारणस्य अभावे, कार्यस्य अभाव: (एव) [vyatireka] - [If kAraNa is not-present, kArya is (certainly) not-present].

This introduction of the word [eva] is game-changer. It now reconciles BhAshyakAra's statement clearly. [Only when] mind is present, there is dvaita. When mind is absent, there is [certainly] no dvaita. If we merely say, if there is mind, there is dvaita - then it is incorrect (because in samAdhi, mind is present but there is no dvaita).

Hope I am able to communicate.

Regards.
Sudhanshu Shekhar.  

Ram Chandran

unread,
Jan 12, 2026, 8:45:41 AM (7 days ago) Jan 12
to advaitin
Namaskar:

Let me clarify my statement regarding explanations provided by AI:  Yes, it is valuable but it may be necessary but not sufficient!! We do need books, Webpages, discussion groups and interactions for getting better information.  No one can deny the valuable contributions through books by our moderators that include Dennis Waite, ProfVK, Acharya Sadanandaji, and others.  I solute them for their contributions,

With my warm regards,
Ram Chandran

Ram Chandran

unread,
Jan 12, 2026, 10:27:43 AM (7 days ago) Jan 12
to advaitin
Namaskar;
The operational scheme of AI is to collect and assemble all pertain information in any subject area.   Any question raised by anyone that includes all discussions and interactions wherever they become available.  In many ways more discussions, thoughts and innovations will enable AI to provide more reliable information to all.  The operation of AI definitely requires appropriate restrictions, regulations to prevent it to go in the wrong direction!
Ram Chandran

Vikram Jagannathan

unread,
Jan 12, 2026, 12:01:38 PM (7 days ago) Jan 12
to adva...@googlegroups.com
Namaskaram Sudhanshu ji,

A very quick comment - if indeed there is the distinction between the anvaya-vyatireka in Advaita (particularly in terms of causality) vs Nyaya; can this is explained on the basis of sat-karya-vada and asat-karya-vada between the two darshanas? This does explain the continued existence of the effect pot in its cause clay.

prostrations,
Vikram


--
You received this message because you are subscribed to the Google Groups "advaitin" group.
To unsubscribe from this group and stop receiving emails from it, send an email to advaitin+u...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/advaitin/CAH9%3D%2BBCXNRs9wo_6n2tz%2BrS5icwHrS4coJH_ekGMLce5bsTgtQ%40mail.gmail.com.

dwa...@advaita.org.uk

unread,
Jan 12, 2026, 12:06:21 PM (7 days ago) Jan 12
to adva...@googlegroups.com

Hi Sudhanshu-ji,

 

That is very apposite and helpful – thank you! (Again I will post this as a comment to the first part, assuming you will again agree to this.)

 

Interestingly, ChatGPT made the same point about the NyAya use of anvaya-vyatireka differing from that of Advaita. But the thrust of my discussion with ChatGPT is not quite on this point but on the implication that, because of the relationships – mind is, world is; mind is not, world is not (in deep sleep) – therefore the world disappears in deep sleep. (You seem to affirm this interpretation in your comments, and of course I know that many agree. With appropriate temerity, I have to say that I don’t agree!) AI then suggests that Śaṅkara’s real intention is a sort of ‘triple anvaya-vyatireka’ to show that the real ‘cause’ of the world (substrate of both mind and world) is Consciousness. And of course That IS always present (hence the world is always present, although of course it is mithyA).

 

It could be that AI is simply trying to corroborate my own views, as it tends to do. The discussion is not yet complete as I aim to track down the supporting references…

 

But I am trying here to summarize the 8000+ words in a few sentences. Please just take it as an indication of the general context of the discussion and await the complete posts.

 

Best wishes,

Dennis

--

Sudhanshu Shekhar

unread,
Jan 12, 2026, 12:38:44 PM (7 days ago) Jan 12
to Advaitin
Namaste Dennis ji. 

//But the thrust of my discussion with ChatGPT is not quite on this point but on the implication that, because of the relationships – mind is, world is; mind is not, world is not (in deep sleep) – therefore the world disappears in deep sleep. (You seem to affirm this interpretation in your comments, and of course I know that many agree. With appropriate temerity, I have to say that I don’t agree!)//

Mere presence of mind does not guarantee presence of dvaita. An example is samAdhi, when mind is, but dvaita is not.

That is where the word "eva" comes in. (Only when) There is mind, there is dvaita. This is what BhAshyakAra intends and this is what Advaita anvaya-vyatireka intends. You will have to agree that "(only when) mind is, dvaita is". The other part, "when mind is not, dvaita is not" is of course agreed by one and all.

This is classic DSV by the way, where entire world is nothing but mind ( = ignorance).

"AI then suggests that Śaṅkara’s real intention is a sort of ‘triple anvaya-vyatireka’ to show that the real ‘cause’ of the world (substrate of both mind and world) is Consciousness. And of course That IS always present (hence the world is always present, although of course it is mithyA)."

This is not the point in 3.31 MK.

Regards.
Sudhanshu Shekhar.

putran M

unread,
Jan 12, 2026, 2:32:02 PM (7 days ago) Jan 12
to adva...@googlegroups.com
Namaskaram Dennis-ji,

AI then suggests that Śaṅkara’s real intention is a sort of ‘triple anvaya-vyatireka’ to show that the real ‘cause’ of the world (substrate of both mind and world) is Consciousness. And of course That IS always present (hence the world is always present, although of course it is mithyA).


I hadn't followed this discussion till now. But Sudhanshu-ji's comment on the above grabbed my attention. The recent talks I was posting was making a similar point as what AI is doing above. 

The world of thoughts ("mind-world") we consider as happening or having locus "in the mind". This we distinguish from the gross world of matter (~ "sense-world") that we consider to exist independently of the mind. In fact (if we go along with physics claims) the matter-world reduces to a substrate "Energy". (I suppose this may be considered as based in SDV framework). 

But we can reduce mind and energy and the worlds they appear as, to objects of knowledge in Consciousness of the Self. 

So, Self ... | Consciousness <- knowledge <- mind, energy <- mind-world, sense-world, etc.

What is the locus of knowledge of the Self? If Self here denotes NB, then the knowledge must be held to be intrinsic to the Consciousness that illumines it and causes/projects the cognition/imagination/appearance of the Self as mind, energy, mind-world, sense-world, etc.

In this case, Consciousness (Maya/Avidya) has the role of both the Mind (knowledge) "of the Self" and the power that illumines this knowledge into cognition ("manifestation"). The subsequent jiva-mind and jagat-energy are thus projections of the primal Mind "of the Self".

(Now the "knowledge" here may be considered ignorance because the content of that knowledge is adhyasa of Self: we are knowing the Self in dualistic terms. Both this dualistic knowledge and its substrate Mind/Maya are mithya.)

Alternatively, we can modify the sequence slightly as:

Self (NB) ... | Self+Consciousness (Sat+Chit, Ishvara) <- knowledge <- mind, energy <- mind-world, sense-world, etc. This makes explicit that Consciousness appears as the shakti "of the Self" and not as if separately; but again, such characterization is also mithya.

thollmelukaalkizhu



dwa...@advaita.org.uk

unread,
Jan 13, 2026, 5:02:17 AM (6 days ago) Jan 13
to adva...@googlegroups.com

Sorry, Putran-ji,

 

I couldn’t follow any of that. Is it symbolic logic you are using? I never did that at school! This is all a minefield of potentially prolonged discussion and I do not want to pursue that.

 

Best wishes,

Dennis

dwa...@advaita.org.uk

unread,
Jan 13, 2026, 5:22:32 AM (6 days ago) Jan 13
to adva...@googlegroups.com

Hi Sudhanshu-ji,

 

You are right (of course!) that GK3.31 does not make the point about the ever-presence of awareness. And ChatGPT made the same point. Its claim was that this was the background inference of Śaṅkara’s argument and referenced Br. Up. Bh. 4.3.23 in support.

 

Best wishes,

Dennis

 

**********************

Michael Chandra Cohen

unread,
Jan 14, 2026, 9:13:17 AM (5 days ago) Jan 14
to adva...@googlegroups.com
Namaste Sudhanshuji et. al. 

I know Sudhanshuji rejects posts/comments that enlist LLMs. I find that short sighted. Artificial Intelligence is a reality we all recognize in our time. I think Dennis's warning to be very careful about what we believe from AI is timely. However, the benefits far outweigh the risks if used correctly.. This is true for any powerful tool we use including our own intellects. Nowhere is this a greater issue than in academia. Accordingly, great universities have formulated policies regarding its use. Here are the policies demanded from established scholars publishing at conferences and in journals.  The bottom line for established researchers: you can typically use AI for editing and polishing, but you must disclose substantive use, cannot list AI as an author or cite it as a source, and remain fully accountable for all content.

Universal prohibition on AI authorship — Elsevier, Springer Nature, Wiley, Taylor & Francis, and SAGE Publishing all explicitly state that generative AI, LLMs, or any similar technologies cannot be listed as an author or co-author. The rationale is that authorship implies responsibilities and tasks that can only be attributed to and performed by humans—each author is accountable for ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

AI cannot be cited as a source — AAAI policy holds that any AI system, including generative models, does not satisfy the criteria for authorship of papers published by AAAI and also cannot be used as a citable source.

Major Conference Policies

AAAI/AIES — Papers that include text generated from an LLM such as ChatGPT are prohibited unless the produced text is presented as part of the paper's experimental analysis. This policy does not prohibit authors from using LLMs for editing or polishing author-written text.

ICML — Authors are allowed to use generative AI tools such as LLMs to assist in writing or research. However, authors must take full responsibility for all content in their paper, including any content generated by AI tools that might be construed as plagiarism or scientific misconduct.

ICML 2026 has added: Any attempts at prompt injection are strictly forbidden and will result in desk rejection (referring to text inserted with the intention to manipulate LLMs to obtain favorable reviews).

IJCAI — Generative AI models, including ChatGPT, BARD, LLaMA, or similar LLMs, do not satisfy the criteria for authorship. If authors use an LLM in any part of the paper-writing process they assume full responsibility for all content, including checking for plagiarism and correctness of all text.

Publisher-Specific Requirements

Springer Nature — Use of AI for minor copy editing (correcting grammar, spelling, punctuation, or tone) does not require disclosure. If AI tools were used to generate content (drafting, rewriting, summarizing), or had any creative/editorial role, disclosure should be made in the Methods section.

Elsevier — Authors who use AI or AI-assisted tools during the manuscript writing process are asked to disclose their use in a separate section of the manuscript, and a statement will appear in the published work. Additionally, you cannot use output from AI agents or deep research directly as the text of your manuscript. AI output can only serve as inspiration in the manuscript preparation process.

Images — Springer Nature does not allow the inclusion of generative AI images in publications. Elsevier similarly prohibits AI-generated or AI-altered images.

Peer Review Restrictions

Policies for reviewers are notably stricter than for authors. No conferences granted reviewers a leniency rating above 3, likely reflecting concerns about the potential risks of using AI tools in the peer-review process, such as the possibility of leaking sensitive or unpublished information. Springer Nature asks peer reviewers not to upload manuscripts into generative AI tools.

Variation by Field

AI conferences, including CVPR, were highly permissive of GenAI use, likely due to greater familiarity with LLMs and their perceived benefits. Interdisciplinary conferences such as UIST and VR had more restrictive policies.


--
You received this message because you are subscribed to the Google Groups "advaitin" group.
To unsubscribe from this group and stop receiving emails from it, send an email to advaitin+u...@googlegroups.com.

Sudhanshu Shekhar

unread,
Jan 16, 2026, 4:24:55 AM (3 days ago) Jan 16
to adva...@googlegroups.com
Namaste Vikram ji.

A very quick comment - if indeed there is the distinction between the anvaya-vyatireka in Advaita (particularly in terms of causality) vs Nyaya; can this is explained on the basis of sat-karya-vada and asat-karya-vada between the two darshanas? This does explain the continued existence of the effect pot in its cause clay.

I applied mind over it. I think the ambit of anvaya-vyatireka is to prove causality. Whether the effect is born from nowhere, or whether the effect was earlier present in cause -- may not be within the domain of anvaya-vyatireka.

I am open to corrections in this regard.

Regards.
Sudhanshu Shekhar.
Reply all
Reply to author
Forward
0 new messages