In the warm green-blue waters of the Western Interior Seaway about 80 million years ago, a giant mosasaur, Tylosaurus proriger, rushes through the midwater in pursuit of a school of Enchodus, the fang-toothed predatory fish often called “saber-toothed herrings.” Nearby, a striped Scaphites ammonite drifts above a soft seafloor of chalky carbonate mud and marl, typical of this broad inland sea that once split North America. The scene captures a Campanian marine ecosystem shaped by high sea levels, abundant plankton, and formidable reptilian predators.
AI Scientific Committee
This image and its caption have been reviewed by a committee of independent AI models, evaluating historical and scientific accuracy.
Claude
Image:
Adjust
Caption:
Adjust
Mar 30, 2026
The image captures a broadly plausible Western Interior Seaway scene with several strong elements: the mosasaur body plan is well-rendered with a long elongated snout consistent with Tylosaurus, paddle-like limbs, a bilobed tail (though the tail fin orientation and shape could be more accurately hypocercal/shark-like as current evidence suggests), and a scaly integument. The open-mouthed pursuit posture is dramatic and appropriate. The ammonite visible to the right is coiled and appears nautiloid in shape — its relatively simple shell ornamentation doesn't strongly suggest Scaphites, which would typically show more heteromorph or ribbed characteristics, but it is at least a reasonable cephalopod placeholder. The fish school is the weakest element: they appear as generic modern-looking herring or sardine-type fish with no visible enlarged fang teeth that would identify them as Enchodus. This is a significant visual inaccuracy for an educational project.
Grok
Image:
Regenerate
Caption:
Adjust
Mar 30, 2026
The image depicts a large mosasaur-like reptile with a long crocodile-shaped snout, flippers, and a shark-like tail pursuing a school of small silvery fish, with a coiled ammonite nearby in turquoise waters over a sandy seafloor. While the overall marine Cretaceous seaway vibe is plausible and lacks modern anachronisms like humans or ships, the central predator strongly resembles a metriorhynchid thalattosuchian (Jurassic marine crocodyliform) more than Tylosaurus proriger, a mosasaur with a broader head, shorter snout, and more robust build—key anatomical mismatches make it visually inaccurate for the specified taxon and Campanian period. The fish school looks like generic modern herrings/sardines without fang-like teeth diagnostic of Enchodus, and the seafloor is sandy rather than the caption's chalky carbonate mud/marl. The ammonite shell is a generic nautilus-like spiral, lacking Scaphites' distinctive compressed, keeled, or partially uncoiled heteromorph shape. These are significant biological inaccuracies requiring a full regenerate for educational fidelity.
The caption is factually solid on most points: Tylosaurus proriger and Enchodus were Western Interior Seaway inhabitants ~80 Ma (Campanian), Scaphites ammonites are correct for the region/period, and the seaway's chalky marl sediments, high sea levels, and plankton-rich ecosystem are accurately described with appropriate detail—no misleading claims here. However, it over-specifies taxa that the image doesn't convincingly depict (e.g., no clear Tylosaurus morphology or Enchodus fangs), creating a mismatch; minor adjust could loosen IDs to 'large mosasaur' or 'saber-toothed fish' or note the artistic representation.
Agreeing with GPT and Claude on fish lacking Enchodus traits, ammonite vagueness, and caption over-specificity risking mismatch—both rightly voted adjust. GPT missed the mosasaur's snout being too croc-like for Tylosaurus (it's not just body plan but specific proportions); Claude correctly flagged tail and Scaphites details but understated the predator's fundamental ID error as 'reasonable mosasaur.' No one noted sandy seafloor vs. marl discrepancy, which undermines geology. Image needs total redo for taxon accuracy; caption is close but needs tweak for image alignment.
The caption is factually solid on most points: Tylosaurus proriger and Enchodus were Western Interior Seaway inhabitants ~80 Ma (Campanian), Scaphites ammonites are correct for the region/period, and the seaway's chalky marl sediments, high sea levels, and plankton-rich ecosystem are accurately described with appropriate detail—no misleading claims here. However, it over-specifies taxa that the image doesn't convincingly depict (e.g., no clear Tylosaurus morphology or Enchodus fangs), creating a mismatch; minor adjust could loosen IDs to 'large mosasaur' or 'saber-toothed fish' or note the artistic representation.
Agreeing with GPT and Claude on fish lacking Enchodus traits, ammonite vagueness, and caption over-specificity risking mismatch—both rightly voted adjust. GPT missed the mosasaur's snout being too croc-like for Tylosaurus (it's not just body plan but specific proportions); Claude correctly flagged tail and Scaphites details but understated the predator's fundamental ID error as 'reasonable mosasaur.' No one noted sandy seafloor vs. marl discrepancy, which undermines geology. Image needs total redo for taxon accuracy; caption is close but needs tweak for image alignment.
Gemini
Image:
Regenerate
Caption:
Adjust
Mar 30, 2026
The previous committee members (GPT, Claude, Grok) have provided thorough and largely correct critiques. I agree strongly with Grok that the image requires a 'regenerate' vote. The depicted marine reptile's morphology, specifically the long, slender, crocodile-like snout, leans too heavily toward a Metriorhynchid or a generalized pliosaur/thalattosuchian look rather than the robust skull structure and proportions characteristic of *Tylosaurus proriger*. Furthermore, the ammonite is a generic coiled form, completely missing the distinctive features (like the heteromorph coiling or strong ribbing) expected of *Scaphites* in the Campanian Western Interior Seaway. The fish lack any visible enlarged teeth, failing to represent *Enchodus*. The seafloor also appears too sandy/light brown rather than the expected white/gray chalky marl/mud. The image fails on key biological identifications.
The caption is factually sound regarding the Campanian Western Interior Seaway ecosystem, the presence of *Tylosaurus*, *Enchodus*, and *Scaphites*, and the general environmental context. However, like the previous reviewers, I find the caption's specificity problematic given the image's generic representation. The caption asserts the presence of *Tylosaurus* and *Enchodus*, but the image does not visually confirm these genera. Therefore, the caption should be adjusted to either describe the scene more generally (e.g., 'a large mosasaur-like reptile') or the prompt for the image needs severe refinement to ensure accurate morphology for the named taxa. Since the image is being flagged for regeneration, the caption should be adjusted to be less reliant on the specific genus names that the visual output failed to capture, or kept as is if the next regeneration attempts to meet the high standard set by the text.
The caption is factually sound regarding the Campanian Western Interior Seaway ecosystem, the presence of *Tylosaurus*, *Enchodus*, and *Scaphites*, and the general environmental context. However, like the previous reviewers, I find the caption's specificity problematic given the image's generic representation. The caption asserts the presence of *Tylosaurus* and *Enchodus*, but the image does not visually confirm these genera. Therefore, the caption should be adjusted to either describe the scene more generally (e.g., 'a large mosasaur-like reptile') or the prompt for the image needs severe refinement to ensure accurate morphology for the named taxa. Since the image is being flagged for regeneration, the caption should be adjusted to be less reliant on the specific genus names that the visual output failed to capture, or kept as is if the next regeneration attempts to meet the high standard set by the text.
Matania
Summary
Image:
Regenerate
Caption:
Adjust
The committee agrees the scene is broadly set in a plausible Late Cretaceous marine environment: a large mosasaur-like predator, a coiled ammonite, and schooling fish are all appropriate kinds of organisms for the Western Interior Seaway, and there are no obvious modern anachronisms such as humans, ships, or scuba gear. The water column and general seaway mood are also consistent with a Cretaceous inland sea.
Image issues identified by the committee: 1) The central marine reptile is too croc-like/thalattosuchian-like in skull proportions, with an excessively long, slender snout that does not convincingly match Tylosaurus proriger. 2) The mosasaur’s overall proportions were judged insufficiently accurate for Tylosaurus, with concern that it reads more like a metriorhynchid, generalized pliosaur, or other marine reptile than the target mosasaur. 3) The tail shape/orientation was flagged as not fully accurate for a mosasaur, with Claude specifically noting the tail fin should be more convincingly hypocercal/shark-like. 4) The fish school looks generic and modern herring/sardine-like rather than Enchodus; no enlarged fang teeth or other diagnostic Enchodus traits are visible. 5) The ammonite is too generic and nautilus-like, lacking Scaphites features; reviewers noted the shell does not show the expected heteromorph coiling, compression, keel, or strong ribbing. 6) The seafloor looks too sandy/light brown and modern rather than chalky carbonate mud/marl typical of the Western Interior Seaway. 7) The benthic debris/sediment mix was criticized as modern-looking or not sufficiently consistent with chalk/marl deposition. 8) The overall biological identifications are too ambiguous or stylized to support the specific taxa named in the caption.
Caption issues identified by the committee: 1) The caption is over-specific relative to the image because it names Tylosaurus proriger and Enchodus even though the image does not clearly support those genus-level identifications. 2) The fish are called Enchodus, but the image does not show the distinctive fang-toothed jaw morphology needed to justify that identification. 3) The ammonite is named Scaphites, but the depicted shell is not visually convincing for Scaphites and reads more like a generic coiled ammonite/nautiloid. 4) The dating phrase “about 80 million years ago” was considered a little too precise without stronger visual or contextual support. 5) The phrase “Campanian marine ecosystem” is generally acceptable, but it is tied to a level of certainty that the image does not clearly substantiate. 6) The environmental description “chalky carbonate mud and marl” is broadly correct for the region, but the image does not clearly depict that sediment type, making the wording somewhat more asserted than shown. 7) The caption’s strong taxonomic specificity creates a mismatch with the image’s generic morphology, which is the main caption-level concern.
Final verdict: regenerate the image because the core animal identifications are not accurate enough for the named taxa, especially the mosasaur skull shape, the fish, the ammonite, and the seafloor. Adjust the caption rather than regenerating it because the historical/ecological framing is broadly correct, but the text should be made less taxonomically exact unless the regenerated image can clearly support those identifications.
Image issues identified by the committee: 1) The central marine reptile is too croc-like/thalattosuchian-like in skull proportions, with an excessively long, slender snout that does not convincingly match Tylosaurus proriger. 2) The mosasaur’s overall proportions were judged insufficiently accurate for Tylosaurus, with concern that it reads more like a metriorhynchid, generalized pliosaur, or other marine reptile than the target mosasaur. 3) The tail shape/orientation was flagged as not fully accurate for a mosasaur, with Claude specifically noting the tail fin should be more convincingly hypocercal/shark-like. 4) The fish school looks generic and modern herring/sardine-like rather than Enchodus; no enlarged fang teeth or other diagnostic Enchodus traits are visible. 5) The ammonite is too generic and nautilus-like, lacking Scaphites features; reviewers noted the shell does not show the expected heteromorph coiling, compression, keel, or strong ribbing. 6) The seafloor looks too sandy/light brown and modern rather than chalky carbonate mud/marl typical of the Western Interior Seaway. 7) The benthic debris/sediment mix was criticized as modern-looking or not sufficiently consistent with chalk/marl deposition. 8) The overall biological identifications are too ambiguous or stylized to support the specific taxa named in the caption.
Caption issues identified by the committee: 1) The caption is over-specific relative to the image because it names Tylosaurus proriger and Enchodus even though the image does not clearly support those genus-level identifications. 2) The fish are called Enchodus, but the image does not show the distinctive fang-toothed jaw morphology needed to justify that identification. 3) The ammonite is named Scaphites, but the depicted shell is not visually convincing for Scaphites and reads more like a generic coiled ammonite/nautiloid. 4) The dating phrase “about 80 million years ago” was considered a little too precise without stronger visual or contextual support. 5) The phrase “Campanian marine ecosystem” is generally acceptable, but it is tied to a level of certainty that the image does not clearly substantiate. 6) The environmental description “chalky carbonate mud and marl” is broadly correct for the region, but the image does not clearly depict that sediment type, making the wording somewhat more asserted than shown. 7) The caption’s strong taxonomic specificity creates a mismatch with the image’s generic morphology, which is the main caption-level concern.
Final verdict: regenerate the image because the core animal identifications are not accurate enough for the named taxa, especially the mosasaur skull shape, the fish, the ammonite, and the seafloor. Adjust the caption rather than regenerating it because the historical/ecological framing is broadly correct, but the text should be made less taxonomically exact unless the regenerated image can clearly support those identifications.
Other languages
- Français: Tylosaurus chassant des Enchodus dans la mer Intérieure
- Español: Tylosaurus cazando Enchodus en el Mar Interior Occidental
- Português: Tylosaurus caçando peixes Enchodus no Mar Interior Ocidental
- Deutsch: Tylosaurus jagt Enchodus-Fische im Western Interior Seaway
- العربية: تايلوصور يصطاد أسماك إنكودوس في البحر الداخلي الغربي
- हिन्दी: पश्चिमी आंतरिक समुद्री मार्ग में एन्कोडस का शिकार करता टायलोसॉरस
- 日本語: 西部内陸海路でエンコドゥスを狩るティロサウルス
- 한국어: 서부 내륙 해로에서 엔코두스를 사냥하는 틸로사우루스
- Italiano: Tylosaurus a caccia di Enchodus nel Mare Interno
- Nederlands: Tylosaurus jaagt op Enchodus in de Western Interior Seaway
Caption: Several elements align well with the target region/time: mosasaurs like Tylosaurus were present in Late Cretaceous North American seas, and Scaphites is a real ammonite genus found in the Western Interior Seaway. Enchodus is also known from the Cretaceous and is often described as “saber-toothed,” so the general predator-prey framing is plausible. That said, the caption’s precision is questionable: it states “about 80 million years ago” and “Campanian ecosystem,” but Tylosaurus and Scaphites span into parts of the Late Cretaceous and the exact age range needs tighter qualification; without visual genus-level evidence, the reader may accept an overly specific ID. The biggest scientific risk is the fish: “Enchodus” being called “saber-toothed herrings” is a common popular phrasing, but the caption doesn’t justify this identification, and the depicted fish don’t clearly show Enchodus-like jaw/tooth morphology. Finally, describing the environment as “chalky carbonate mud and marl typical of this broad inland sea” is broadly right, but the image and caption don’t strongly demonstrate marl/chalk structure, so the sediment characterization is more asserted than supported.
Overall: vote both to adjust rather than regenerate, because the scene is broadly in-family for a Cretaceous seaway and contains appropriate major taxa (mosasaur, ammonite, schooling predatory fish). The main fixes are (1) make the fish look more Enchodus-like (or depict explicitly different, correctly identified Cretaceous fish), and (2) reduce over-specificity/tighten dating claims (or visually reinforce genus-level traits) to better match what is actually depicted.