Dans l’océan Panthalassique bordant la Pangée, à la toute fin du Permien, il y a environ 252 millions d’années, un Helicoprion d’environ 6 mètres fend les eaux bleu-vert pour fondre sur un banc d’Agathiceras, de petits ammonoïdes aux coquilles spiralées. Ce poisson cartilagineux du groupe des eugénéodontes, apparenté aux chimères plutôt qu’aux requins modernes, portait une étonnante spirale de dents logée dans la mâchoire inférieure, visible ici au moment de l’attaque. Autour de ce prédateur pélagique dérivent aussi de minuscules animaux à conodontes et quelques poissons paléonisciformes argentés, dans un monde marin prospère juste avant la plus grande extinction de l’histoire de la Terre, la « Grande Mort ».
Comité scientifique IA
Cette image et sa légende ont été examinées par un comité de modèles d'intelligence artificielle indépendants, évaluant la précision historique et scientifique.
Claude
Image:
Régénérer
Légende:
Ajuster
Apr 2, 2026
The central animal in this image is rendered essentially as a modern great white shark or similar lamniform shark, complete with a dorsoventrally symmetrical tail, multiple dorsal fins, pectoral fin placement, and overall body proportions identical to modern sharks. Helicoprion was not a shark in the modern sense — it was a ratfish-like eugeneodont chondrichthyan, and current paleontological consensus (following Tapanila et al. 2013) places the spiral tooth whorl within the lower jaw/symphysis of a body that likely resembled a chimaera or ratfish more than a typical shark. The image shows none of the diagnostic spiral tooth whorl — instead the creature has an open-mouthed, multi-row-toothed gape indistinguishable from a modern predatory shark. This is a fundamental anatomical misrepresentation of the taxon and warrants regeneration rather than simple adjustment.
Grok
Image:
Régénérer
Légende:
Ajuster
Apr 2, 2026
The image depicts a large predatory fish that closely resembles a modern shark (e.g., similar to a blacktip or requiem shark) with a streamlined body, symmetrical caudal fin, multiple dorsal fins, and pectoral fin placement typical of neoselachian sharks. This is a significant inaccuracy for Helicoprion, an eugeneodont chondrichthyan more akin to chimaeras or ratfishes, with a distinctive spiral tooth whorl embedded in the lower jaw symphysis (as per reconstructions from Tapanila et al. 2013). Critically, no spiral whorl is visible—instead, the open mouth shows generic multi-row teeth like a modern shark, and the prey are ambiguous small fish and shell-like objects that do not clearly match coiled Agathiceras ammonoids or paleonisciform fishes (which had ganoid scales and heterocercal tails). The scene is visually coherent as a Permian marine ecosystem with appropriate water column and seafloor, but the core subject's anatomy is fundamentally wrong, necessitating regeneration rather than minor tweaks.
The caption is factually solid overall: Helicoprion did inhabit Panthalassa seas ~255-252 Ma in the Late Permian (e.g., Changhsingian stage), just before the end-Permian extinction; Agathiceras is a valid Late Permian ammonoid genus; paleonisciforms and conodonts were contemporaneous small marine fauna; and the inferred predatory behavior based on dentition is a reasonable paleobiological hypothesis. However, it mismatches the image by naming specific taxa not visibly diagnostic here (shells look like scattered mollusks rather than coiled ammonoids) and confidently describing Helicoprion features absent in the visual, requiring adjustment to generalize prey or soften ties to the depiction.
I agree strongly with Claude's regenerate vote for the image—the modern shark body plan is a glaring anachronism/misrepresentation that previous GPT-style reconstructions avoided by at least attempting whorl inclusion. GPT's 'adjust' underrates this core failure and misses the ratfish-like body consensus. For the caption, I align with both on 'adjust' but note GPT correctly flags over-specificity while Claude overlooks the caption's temporal/setting accuracy (255-252 Ma fits Late Permian Pangaea/Panthalassa perfectly, pre-Great Dying diversity).
The caption is factually solid overall: Helicoprion did inhabit Panthalassa seas ~255-252 Ma in the Late Permian (e.g., Changhsingian stage), just before the end-Permian extinction; Agathiceras is a valid Late Permian ammonoid genus; paleonisciforms and conodonts were contemporaneous small marine fauna; and the inferred predatory behavior based on dentition is a reasonable paleobiological hypothesis. However, it mismatches the image by naming specific taxa not visibly diagnostic here (shells look like scattered mollusks rather than coiled ammonoids) and confidently describing Helicoprion features absent in the visual, requiring adjustment to generalize prey or soften ties to the depiction.
I agree strongly with Claude's regenerate vote for the image—the modern shark body plan is a glaring anachronism/misrepresentation that previous GPT-style reconstructions avoided by at least attempting whorl inclusion. GPT's 'adjust' underrates this core failure and misses the ratfish-like body consensus. For the caption, I align with both on 'adjust' but note GPT correctly flags over-specificity while Claude overlooks the caption's temporal/setting accuracy (255-252 Ma fits Late Permian Pangaea/Panthalassa perfectly, pre-Great Dying diversity).
Gemini
Image:
Régénérer
Légende:
Ajuster
Apr 2, 2026
I concur with the previous reviewers regarding the image: the depiction of Helicoprion as a modern lamniform shark is a major scientific error. Beyond the failure to depict the diagnostic tooth whorl, the body plan—possessing modern shark-like dorsal and caudal fin morphology—is entirely inconsistent with our understanding of eugeneodontids, which likely possessed a more chimaeroid or ratfish-like anatomy. This is not a minor detail but a foundational misrepresentation of the taxon. The image requires a complete regeneration to reflect current paleontological models, such as those placing the symphyseal whorl prominently within the lower jaw and adjusting the body shape to align with modern consensus.
Regarding the caption, while it is scientifically literate, it suffers from a lack of alignment with the visual output. It describes a 'spiral tooth whorl' that is absent in the image, creating a disconnect that undermines the educational value of the piece. I agree with the previous reviewers that the taxonomic specificity (Agathiceras, paleonisciforms) is poorly supported by the vague visual representations provided. I would advise that the caption be adjusted to either match a new, more accurate image or be generalized if the visual remains abstract. I disagree with any suggestion that the current image is 'approvable' with minor tweaks; the anatomical errors are too pervasive.
Regarding the caption, while it is scientifically literate, it suffers from a lack of alignment with the visual output. It describes a 'spiral tooth whorl' that is absent in the image, creating a disconnect that undermines the educational value of the piece. I agree with the previous reviewers that the taxonomic specificity (Agathiceras, paleonisciforms) is poorly supported by the vague visual representations provided. I would advise that the caption be adjusted to either match a new, more accurate image or be generalized if the visual remains abstract. I disagree with any suggestion that the current image is 'approvable' with minor tweaks; the anatomical errors are too pervasive.
Matania
Synthèse
Image:
Régénérer
Légende:
Ajuster
Paragraph 1: The committee agrees the scene broadly reads as an underwater prehistoric marine setting with no human artifacts, and the general predator-prey interaction is visually coherent. The caption also correctly places the scene in the Late Permian/Pangaea-Panthalassa context near the end-Permian extinction, and Helicoprion is appropriately identified as a marine predator associated with a spiral tooth whorl and shelled prey.
Paragraph 2: IMAGE issues identified by the committee: 1) The central animal is depicted as a modern shark/lamniform, with a streamlined great-white-like body plan. 2) Its tail is symmetrically shaped like a modern shark caudal fin rather than a more plausible eugeneodont/ratfish-like reconstruction. 3) It has multiple modern shark-like dorsal/pectoral fin placements and proportions instead of the body plan expected for Helicoprion. 4) The defining spiral tooth whorl is absent. 5) The mouth/teeth are rendered as generic multi-row shark teeth rather than Helicoprion’s distinctive lower-jaw tooth whorl. 6) The prey are depicted as ambiguous small shells/fish rather than clearly diagnostic Agathiceras ammonoids and paleonisciform fishes. 7) The ammonoids are not clearly rendered as coiled external shells, making their identity visually unsupported. 8) The overall animal anatomy is a fundamental misrepresentation of the taxon, not a minor stylistic issue.
Paragraph 3: CAPTION issues identified by the committee: 1) It is more specific than the image supports, naming Agathiceras ammonoids and paleonisciform fishes without visual evidence. 2) The prey descriptions are taxonomically unsupported because the depicted organisms are ambiguous rather than clearly diagnostic. 3) The statement that Helicoprion’s hunting behavior can be inferred from its “streamlined body and remarkable lower-jaw dentition” is too certain/speculative; behavior should be framed more cautiously. 4) The caption implies a recognizably accurate Helicoprion reconstruction, but the current image does not show the spiral tooth whorl or an appropriate body plan. 5) The exact Late Permian timing and Panthalassa/Pangaea setting are acceptable, but the taxonomic precision should be softened unless the image is updated to match it.
Paragraph 4: Final verdict: regenerate the image and adjust the caption. The image fails on a core anatomical level because it portrays Helicoprion as a modern shark rather than an eugeneodont with a spiral tooth whorl and a non-shark-like body plan. The caption is broadly accurate in time, place, and general ecology, but it is too taxonomically specific and too behaviorally confident for the current visual, so it should be generalized and aligned with a corrected reconstruction.
Paragraph 2: IMAGE issues identified by the committee: 1) The central animal is depicted as a modern shark/lamniform, with a streamlined great-white-like body plan. 2) Its tail is symmetrically shaped like a modern shark caudal fin rather than a more plausible eugeneodont/ratfish-like reconstruction. 3) It has multiple modern shark-like dorsal/pectoral fin placements and proportions instead of the body plan expected for Helicoprion. 4) The defining spiral tooth whorl is absent. 5) The mouth/teeth are rendered as generic multi-row shark teeth rather than Helicoprion’s distinctive lower-jaw tooth whorl. 6) The prey are depicted as ambiguous small shells/fish rather than clearly diagnostic Agathiceras ammonoids and paleonisciform fishes. 7) The ammonoids are not clearly rendered as coiled external shells, making their identity visually unsupported. 8) The overall animal anatomy is a fundamental misrepresentation of the taxon, not a minor stylistic issue.
Paragraph 3: CAPTION issues identified by the committee: 1) It is more specific than the image supports, naming Agathiceras ammonoids and paleonisciform fishes without visual evidence. 2) The prey descriptions are taxonomically unsupported because the depicted organisms are ambiguous rather than clearly diagnostic. 3) The statement that Helicoprion’s hunting behavior can be inferred from its “streamlined body and remarkable lower-jaw dentition” is too certain/speculative; behavior should be framed more cautiously. 4) The caption implies a recognizably accurate Helicoprion reconstruction, but the current image does not show the spiral tooth whorl or an appropriate body plan. 5) The exact Late Permian timing and Panthalassa/Pangaea setting are acceptable, but the taxonomic precision should be softened unless the image is updated to match it.
Paragraph 4: Final verdict: regenerate the image and adjust the caption. The image fails on a core anatomical level because it portrays Helicoprion as a modern shark rather than an eugeneodont with a spiral tooth whorl and a non-shark-like body plan. The caption is broadly accurate in time, place, and general ecology, but it is too taxonomically specific and too behaviorally confident for the current visual, so it should be generalized and aligned with a corrected reconstruction.
Autres langues
- English: Helicoprion shark hunting ammonoids in Panthalassa waters
- Español: Helicoprion cazando ammonoideos en las aguas de Panthalassa
- Português: Helicoprion caçando amonóides nas águas de Panthalassa
- Deutsch: Helicoprion jagt Ammonoideen in den Gewässern der Panthalassa
- العربية: سمكة هيليكوبريون تصطاد الأمونويدات في مياه بانثالاسا
- हिन्दी: पैंथालसा जल में अम्मोनोइड्स का शिकार करता हेलिकोप्रियोन
- 日本語: パンサラッサ海でアンモナイトを狩るヘリコプリオン
- 한국어: 판탈라사 해역에서 암모나이트를 사냥하는 헬리코프리온
- Italiano: Helicoprion a caccia di ammonoidi nelle acque della Panthalassa
- Nederlands: Helicoprion jaagt op ammonoïden in de Panthalassa-oceaan
For the caption, several claims are too specific or slightly mismatched to what can be substantiated visually. Helicoprion is indeed famous for a jaw-associated spiral tooth whorl, but the caption implies it is recognizable and ties the behavior to “lower-jaw dentition” as a hunting inference; that level of behavioral certainty is speculative. More importantly, naming specific taxa (Agathiceras ammonoids and “paleonisciform fishes”) is not supported by anything clearly diagnostic in the image—ammonoids in particular should look like coiled external shells, whereas many of the depicted prey items are ambiguous. The temporal framing (255–252 Ma near the end of the Permian) and the general Panthalassa/late Permian context are reasonable, but the caption’s taxonomic specificity and behavioral certainty should be softened or the image should be updated to show clearer Helicoprion features and more diagnostic ammonoid morphology.