後期ジュラ紀、およそ1億600万〜1億450万年前のテチス海の礁頂では、澄んだ浅いターコイズ色の海を波が洗い、ドーム状のサンゴ *Isastrea* と板状の *Thamnasteria* が、花瓶形のケイ質海綿 *Cnemidiastrum* と複雑に組み合って丈夫な炭酸塩リーフを築いていました。足元の白っぽい炭酸塩砂や硬い海底には、殻径5〜10 cmほどのウニ *Cidaris* が棘を動かしながら這い、銀色にきらめく小魚 *Leptolepis* の群れがサンゴの間をすり抜けます。これは、現代のサンゴ礁とは少し異なり、造礁サンゴと海綿がともに主役を担っていたジュラ紀の温暖浅海生態系を描いた情景です。
AI科学委員会
この画像とキャプションは、歴史的・科学的正確性を評価する独立したAIモデルの委員会によってレビューされました。
Claude
画像:
調整
キャプション:
調整
Apr 2, 2026
The image presents a visually appealing and morphologically diverse shallow tropical reef scene with dome-shaped coral masses, vase/cup-shaped sponge-like forms, plate-like structures, sea urchins, and small fish—all broadly consistent with a Late Jurassic warm Tethys carbonate reef. The varied morphotypes are actually a better match to the named taxa (Isastrea, Thamnasteria, Cnemidiastrum) than the previous reviewer credited. However, the sea urchins are the clearest inaccuracy: their spines appear modern in form—long, slender, and needle-like—whereas Jurassic cidaroids like Cidaris had distinctly thick, club-shaped primary spines. This is a scientifically significant misrepresentation that warrants correction. The fish are generically acceptable as Leptolepis-type primitives. The caption's core claims are scientifically sound—all named taxa are documented from the Jurassic Tethys, the date range of 160–150 Ma is accurate, and the carbonate platform description is well-grounded. I disagree with GPT's suggestion that naming specific genera is problematic; representative taxa are standard in paleontological reconstruction captions. A minor improvement would be adding a clarifying phrase that these are representative reconstructed organisms. The closing educational contrast about modern reef fish and flowering plants is appropriate and useful rather than distracting. Primary action item: regenerate sea urchins with cidaroid-appropriate spine morphology.
Grok
画像:
調整
キャプション:
承認
Apr 2, 2026
The image depicts a vibrant, shallow turquoise reef scene with dome-shaped and bulbous coral-like mounds (plausibly matching Isastrea), fan/plate-like encrustations (evocative of Thamnasteria), vase/cup-shaped sponges (fitting Cnemidiastrum), numerous sea urchins, small silvery fish darting above, and sandy carbonate hardground—all visually coherent and plausible for a Late Jurassic Tethys reef crest. The high-energy clear-water platform is well-evoked with waves and sunlight. No anachronisms like modern fish or plants are present. However, the sea urchins are a key issue: while Cidaris is correct for the period, their rendered spines are long, thin, and needle-like (resembling modern echinoids), not the characteristic short, thick, club-shaped primary spines of Jurassic cidaroids, which is a scientifically noticeable inaccuracy requiring prompt refinement for visual accuracy. Fish are generically appropriate as Leptolepis-like teleosts.
The caption is factually precise: Late Jurassic (160–150 Ma) Tethys reefs were indeed built by such corals (Isastrea, Thamnasteria) and siliceous sponges (Cnemidiastrum), with Cidaris urchins and Leptolepis fish documented from these environments; the high-energy carbonate platform description aligns with Oxfordian-Kimmeridgian geology, and the contrast with modern biota adds useful educational context without misleading. Detail level is appropriate and consistent with the scene.
Agreeing with both GPT and Claude on the sea urchin spine morphology as the primary flaw (Claude's point is spot-on and decisive for 'adjust'), but I find the image's morphotype diversity stronger than GPT credited—vases and plates are clearly present, justifying the named taxa better than ambiguous 'boulder clasts.' GPT underplayed visual taxon matches; Claude missed no major issues but rightly prioritizes cidaroid spines. Caption merits approval over GPT's nitpicks, as specific genera are standard and accurate in paleoart.
The caption is factually precise: Late Jurassic (160–150 Ma) Tethys reefs were indeed built by such corals (Isastrea, Thamnasteria) and siliceous sponges (Cnemidiastrum), with Cidaris urchins and Leptolepis fish documented from these environments; the high-energy carbonate platform description aligns with Oxfordian-Kimmeridgian geology, and the contrast with modern biota adds useful educational context without misleading. Detail level is appropriate and consistent with the scene.
Agreeing with both GPT and Claude on the sea urchin spine morphology as the primary flaw (Claude's point is spot-on and decisive for 'adjust'), but I find the image's morphotype diversity stronger than GPT credited—vases and plates are clearly present, justifying the named taxa better than ambiguous 'boulder clasts.' GPT underplayed visual taxon matches; Claude missed no major issues but rightly prioritizes cidaroid spines. Caption merits approval over GPT's nitpicks, as specific genera are standard and accurate in paleoart.
Gemini
画像:
調整
キャプション:
承認
Apr 2, 2026
I concur with the previous reviewers regarding the sea urchin morphology. While the scene is compositionally excellent and captures the ecological essence of a Jurassic Tethyan carbonate platform, the depiction of Cidaris-type urchins with modern, needle-like spines is a notable paleontological inaccuracy. Jurassic cidaroids are characterized by their robust, club-like or paddle-shaped primary spines, which are distinct from the slender, flexible spines of extant echinoids. A minor adjustment to the AI's rendering of these spines would significantly increase the scientific rigor of the image. Regarding the caption, I disagree with GPT's critique; naming specific, period-appropriate genera is standard practice in educational paleo-reconstruction and provides necessary depth. The caption is factually sound and provides excellent context. I agree with Claude and Grok that the image successfully depicts the requested diversity of growth forms (vases, plates, domes), making the visual support for the caption's taxon claims sufficient once the urchin detail is corrected.
Matania
要約
画像:
調整
キャプション:
調整
The committee agrees the scene is broadly accurate for a Late Jurassic tropical Tethyan shallow-marine carbonate reef: the turquoise clear water, reef-crest setting, mound/plate/vase-like framework, small fish, and overall warm-seas ecology are all consistent with the period. The caption’s basic time/place framing and the use of representative Jurassic taxa are also generally supported. However, the image contains a significant morphological issue with the sea urchins, and the caption makes several taxonomic identifications more specific than the image can clearly support.
IMAGE issues identified by the committee: 1) The sea urchins are rendered with long, thin, needle-like spines that read as modern echinoids rather than Jurassic cidaroids; Jurassic Cidaris should have thicker, stouter, club-shaped primary spines. 2) The fish are generic and do not clearly read as Leptolepis specifically. 3) GPT noted that the dominant dome-shaped coral look is visually ambiguous and can read as generic carbonate clasts, calcareous sponge forms, or microbial mounds rather than clearly diagnostic Jurassic coral morphologies. 4) GPT also noted the scene does not clearly show distinct plate-like Thamnasteria or vase-like sponge morphologies strongly enough to justify the caption’s exact taxa. 5) Claude and Grok disagreed with the severity of the taxon-shape ambiguity, but still agreed the urchin morphology is the primary scientific error.
CAPTION issues identified by the committee: 1) The caption names exact genera/species-like taxa whose diagnostic forms are not unambiguously visible in the image: Isastrea, Thamnasteria, Cnemidiastrum, Cidaris, and Leptolepis. 2) GPT specifically judged the taxonomic claims overconfident because the image does not clearly support plate-like Thamnasteria or vase-shaped Cnemidiastrum, and the fish are not visibly diagnostic of Leptolepis. 3) GPT suggested the phrase ‘high-energy, clear-water carbonate platforms’ is plausible but interpretive; a more general reef/carbonate-platform description would be safer. 4) GPT considered the closing contrast with ‘modern reef fish and flowering plants’ true but somewhat unnecessary and slightly distracting from the paleoecology. 5) Claude, Grok, and Gemini defended the caption as broadly accurate and standard for reconstruction captions, so these are mainly issues of evidentiary certainty rather than outright factual errors.
Verdict: adjust for both image and caption. The image needs a concrete anatomical correction to the sea urchins, which is a clear scientific inaccuracy. The caption is mostly sound but should be softened where it over-specifies taxa and habitat details that the image does not conclusively demonstrate. Because the main remaining problems are specific and fixable, adjustment is sufficient rather than regeneration.
IMAGE issues identified by the committee: 1) The sea urchins are rendered with long, thin, needle-like spines that read as modern echinoids rather than Jurassic cidaroids; Jurassic Cidaris should have thicker, stouter, club-shaped primary spines. 2) The fish are generic and do not clearly read as Leptolepis specifically. 3) GPT noted that the dominant dome-shaped coral look is visually ambiguous and can read as generic carbonate clasts, calcareous sponge forms, or microbial mounds rather than clearly diagnostic Jurassic coral morphologies. 4) GPT also noted the scene does not clearly show distinct plate-like Thamnasteria or vase-like sponge morphologies strongly enough to justify the caption’s exact taxa. 5) Claude and Grok disagreed with the severity of the taxon-shape ambiguity, but still agreed the urchin morphology is the primary scientific error.
CAPTION issues identified by the committee: 1) The caption names exact genera/species-like taxa whose diagnostic forms are not unambiguously visible in the image: Isastrea, Thamnasteria, Cnemidiastrum, Cidaris, and Leptolepis. 2) GPT specifically judged the taxonomic claims overconfident because the image does not clearly support plate-like Thamnasteria or vase-shaped Cnemidiastrum, and the fish are not visibly diagnostic of Leptolepis. 3) GPT suggested the phrase ‘high-energy, clear-water carbonate platforms’ is plausible but interpretive; a more general reef/carbonate-platform description would be safer. 4) GPT considered the closing contrast with ‘modern reef fish and flowering plants’ true but somewhat unnecessary and slightly distracting from the paleoecology. 5) Claude, Grok, and Gemini defended the caption as broadly accurate and standard for reconstruction captions, so these are mainly issues of evidentiary certainty rather than outright factual errors.
Verdict: adjust for both image and caption. The image needs a concrete anatomical correction to the sea urchins, which is a clear scientific inaccuracy. The caption is mostly sound but should be softened where it over-specifies taxa and habitat details that the image does not conclusively demonstrate. Because the main remaining problems are specific and fixable, adjustment is sufficient rather than regeneration.
Other languages
- English: Jurassic Tethys Reef Crest with Isastrea Corals and Sponges
- Français: Crête de récif de la Téthys avec coraux Isastrea
- Español: Arrecife del Tetis jurásico con corales Isastrea y esponjas
- Português: Recife de Tétis jurássico com corais Isastrea e esponjas
- Deutsch: Jurassisches Tethys-Riff mit Isastrea-Korallen und Kieselschwämmen
- العربية: شعاب بحر تيثيس الجوراسية مع مرجان إيزاستريا وإسفنج
- हिन्दी: टेथिस सागर की जुरासिक प्रवाल भित्तियाँ और सिलिसियस स्पंज
- 한국어: 테티스해의 쥐라기 산호초와 이사스트레아 산호 군락
- Italiano: Barriera corallina della Tetide giurassica con coralli Isastrea
- Nederlands: Jurassisch Tethys-rif met Isastrea-koralen en kiezelsponzen
However, the dominant “dome-shaped coral” look appears visually ambiguous: the large, smooth, boulder-like forms read more like generic carbonate clasts or calcareous sponge/microbial mound analogs than specific Jurassic coral or sponge morphotypes. The sea urchins are plausible, but their abundance and coloration look modern/graphic rather than fossil-fauna rendering; also, the fish look generic and the caption-specific identification (Leptolepis) is not visually supported. Additionally, no distinct plate-like or vase-like sponge morphologies are clearly represented—most encrustations are clustered as rounded lobes, which makes the named taxa hard to justify from the image alone.
For the caption, the time/place framing (Late Jurassic, tropical Tethys, warm shallow waters, ~160–150 Ma) is reasonable. But several taxon claims are likely overconfident: Isastrea, Thamnasteria, Cnemidiastrum, Cidaris, and Leptolepis can all be reasonable Jurassic-era examples in general, yet the image doesn’t clearly depict the diagnostic forms needed to support those exact identifications (e.g., plate-like Thamnasteria, vase-shaped siliceous sponges). The statement “high-energy, clear-water carbonate platforms” is plausible, but the reef crest wave/clarity cues are interpretive; given the image’s more mound-like framework, it would be safer to describe the habitat more generally as a carbonate reef/complex rather than asserting specific construction by those exact genera. Finally, “long before modern reef fish and flowering plants appeared” is true but unnecessary and slightly shifts the educational focus away from the core paleoecology.