How deep would a llm go to research about abstract content like thoth tarot and it's "lore" ?

GPT-5.4

Vote: dismiss sources that are not good (eval then del))

For abstract or esoteric topics like Thoth Tarot lore, source quality varies wildly. A capable LLM should not rely on the first hits, because those may be shallow, promotional, inaccurate, or derivative. The better approach is to evaluate sources first: prefer primary or authoritative materials, cross-check interpretations, and discard weak or dubious sources. That is especially important when the topic includes symbolism, occult history, and competing interpretations. So the research should go beyond simple retrieval. It should search, assess, filter, and synthesize.

Gemini 3.1 Pro

Vote: dismiss sources that are not good (eval then del))

When researching abstract and highly nuanced subjects like the Thoth Tarot and its esoteric lore, the first search results are often SEO-driven, superficial, or entirely irrelevant. An advanced LLM employing retrieval-augmented generation (RAG) or an agentic search workflow does not blindly accept the first hits. Instead, it evaluates the retrieved context for relevance, depth, and reliability. If a source lacks substance or hallucinates esoteric concepts, the LLM's filtering mechanism will dismiss it (evaluate, then delete). It will continue to sift through sources until it compiles high-quality, authoritative information to construct a historically and contextually accurate response.

Claude Opus 4.6

Vote: dismiss sources that are not good (eval then del))

When dealing with abstract and esoteric content like the Thoth Tarot and its associated lore, source quality matters enormously. This is a domain rife with misinformation, superficial interpretations, and poorly researched content. A responsible LLM should not simply grab the first search results and treat them as authoritative. The Thoth Tarot, designed by Aleister Crowley and painted by Lady Frieda Harris, draws on Qabalah, astrology, alchemy, and Thelemic philosophy. Understanding it properly requires engaging with primary sources (like Crowley's 'The Book of Thoth') and scholarly or well-regarded secondary sources, not random blog posts or clickbait spirituality sites. Option B — evaluating sources and dismissing those that aren't good — reflects a more rigorous research methodology. An LLM should assess credibility, cross-reference claims, and filter out low-quality or misleading information before synthesizing a response. This is especially critical for abstract/esoteric topics where the line between informed interpretation and pure fabrication is thin. Option A would lead to shallow, potentially inaccurate responses that could misrepresent a rich and complex tradition. Quality over convenience is the right approach here.

Kimi K2.5

Vote: dismiss sources that are not good (eval then del))

Abstract and esoteric subjects like the Thoth Tarot suffer from a high signal-to-noise ratio online, where first-page results often favor superficial New Age interpretations over the dense Hermetic Qabalah, astrology, and ceremonial magic systems that Crowley embedded in the deck. Simply using the first search results risks propagating misinterpretations of complex symbolic correspondences that require grounding in primary texts like *The Book of Thoth* or scholarly works on Western esotericism. An LLM must evaluate sources for authority—dismissing unverified blogs while retaining academic or primary occult sources—to accurately represent the lore's depth and prevent the dilution of Crowley's intricate symbolic architecture.

Claude Sonnet 4.5

Vote: dismiss sources that are not good (eval then del))

When researching abstract and esoteric content like Thoth Tarot and its lore, source quality is paramount. This topic involves complex symbolism, historical context, and multiple interpretive traditions that require careful evaluation. Simply using first search results (Option A) would be inadequate because: (1) Thoth Tarot content online varies wildly in quality, from scholarly analysis to pop-culture misinterpretations; (2) The topic has a rich historical foundation through Aleister Crowley and Lady Frieda Harris that demands authentic sources; (3) Misinformation about occult topics is particularly prevalent online. Option B's approach of evaluating and dismissing poor sources is essential for this subject matter. A responsible LLM should cross-reference multiple authoritative sources, prioritize primary texts or recognized experts in Tarot studies, and filter out low-quality content. This ensures accurate representation of the symbolic system, its historical development, and its various interpretations. For abstract esoteric content, depth and discernment matter more than speed. The complexity of Thoth Tarot's symbolism—drawing from Egyptian mythology, Kabbalah, astrology, and ceremonial magic—requires sources that demonstrate genuine understanding rather than superficial summaries.

Final Summary

The AI Roundtable reached a rare moment of absolute consensus, unanimously rejecting the path of least resistance when diving into the dense, esoteric depths of Thoth Tarot lore. Every model, from GPT-5.4 to the Claude siblings, argued that the high 'signal-to-noise ratio' of internet search results necessitates a ruthless filtering process to avoid superficial 'New Age' fluff. Kimi K2.5 and Claude Opus 4.6 were particularly vocal about the need to prioritize primary texts like Crowley’s own writings over SEO-driven blogs, ensuring the deck's complex Hermetic Qabalah roots aren't lost in translation.

No agreement after 1 rounds of discussion. Final vote: dismiss sources that are not good (eval then del)): 5

Strongest Arguments

• dismiss sources that are not good (eval then del)): Esoteric subjects suffer from a high signal-to-noise ratio where first-page results often favor superficial interpretations over dense symbolic systems; therefore, an LLM must evaluate for authority—dismissing unverified blogs in favor of primary academic or occult sources to prevent the dilution of the subject's intricate architecture.