{"version":1,"type":"story","url":"https://digestai.news/story/gpt-4o-invents-answers-claude-keeps-answering-b","json":"https://digestai.news/story/gpt-4o-invents-answers-claude-keeps-answering-b.json","markdown":"https://digestai.news/story/gpt-4o-invents-answers-claude-keeps-answering-b.md","slug":"gpt-4o-invents-answers-claude-keeps-answering-b","headline":"GPT-4o invents answers, Claude keeps answering b","summary":"The author tested three vision AI models on a set of images designed to be visually appealing but ambiguous. The models were asked to generate captions or answers, and the author found that they largely ignored the visual content, focusing instead on textual cues.\n\nThese observations highlight how different AI systems handle ambiguous prompts and the importance of evaluating model behavior beyond surface-level accuracy. The experiment underscores the need for robust testing frameworks to detect unintended model responses.","keyPoints":["Three vision AI models ignored the pretty image.","GPT-4o invented answers referencing the image.","Claude kept answering with the letter 'B'."],"whyItMatters":"Shows how models differ in handling ambiguous prompts, stressing the need for careful evaluation.","category":{"slug":"models","name":"Generative AI & Models","url":"https://digestai.news/category/models"},"entities":{"companies":[],"models":["GPT-4o","Claude"],"people":[]},"firstPublishedAt":"2026-10-12T11:00:00Z","updatedAt":"2026-10-12T11:00:00Z","sourceCount":1,"hasPrimarySource":false,"sources":[{"outlet":"Towards Data Science","title":"I Built a Trap for AI Image Judges. They Walked Into a Different One.","url":"https://towardsdatascience.com/i-built-a-trap-for-ai-image-judges-they-walked-into-a-different-one","publishedAt":"2026-10-12T11:00:00Z","type":"newsletter","primary":false,"lead":true}],"sourceNotes":null,"discussions":[],"thread":null,"cite":{"text":"Digest AI, \"GPT-4o invents answers, Claude keeps answering b\", 12 October 2026, https://digestai.news/story/gpt-4o-invents-answers-claude-keeps-answering-b","publisher":"Digest AI","title":"GPT-4o invents answers, Claude keeps answering b","datePublished":"2026-10-12T11:00:00Z","url":"https://digestai.news/story/gpt-4o-invents-answers-claude-keeps-answering-b"},"generatedBy":"Written by Digest AI's editorial model from the linked sources; the sources are the record.","license":"Headlines, digests and key points are written by Digest AI and may be quoted with a link to the story page. Linked articles belong to their publishers. Terms: https://digestai.news/terms#reuse"}