{"id":44799,"topic":"ai","source":"6abc Philadelphia","title":"Study examines AI chat bot bias, how it responds to prompts - 6abc Philadelphia","url":"https://6abc.com/post/study-examines-ai-chat-bot-bias-how-responds-prompts/19564010/","url_hash":"c5bb8a076215ccfdf2cb75c39324669e37205100","author":"","summary":"Study examines AI chat bot bias, how it responds to prompts 6abc Philadelphia","content":"PHILADELPHIA (WPVI) -- A new study highlighted by The Washington Post is examining whether artificial intelligence chatbots provide factual answers or reflect political bias when responding to users' questions.\nResearchers evaluated several AI tools, including ChatGPT, Gemini and Claude, using a range of benchmarks and political prompts to compare how the systems responded.\nAccording to the study, Gemini scored the highest for presenting multiple viewpoints, with 93% of responses laying out both sides of an issue.\nClaude maintained neutrality more than half of the time, at 57%.\nChatGPT was found to be left-leaning 80% of the time.\nThe models were instructed to answer questions in 30 words and were tested without personalization settings.\nPrompts covered topics including the death penalty and gun control.\nUniversity of Pennsylvania professor Chris Callison-Burch demonstrated responses from Claude and ChatGPT using a prompt about affirmative action.\n\"Here it's saying this is a contested policy question, so it's saying it has multiple perspectives, not just one,\" Callison-Burch said.\nHe noted similarities in how the chatbots approached the question.\n\"Both Claude here and ChatGPT both seem to try to do this,\" he said.\nCallison-Burch said AI chatbots are trained on vast amounts of internet data and can reflect the behavior and preferences of their users.\n\"It understands like what you read and who you are personally. So it tries to adjust itself to be more helpful for you,\" he said. \"So if you're a Republican, it all tends to nudge itself in that direction. If you're a Democrat, nudge itself in that direction.\"\nDespite concerns about bias, Callison-Burch said there is no need for alarm because people ultimately make their own decisions rather than relying on a chatbot to decide for them.\nHe added that users should remain aware of AI's limitations as more people turn to the technology as a resource.","image_url":"https://cdn.abcotvs.com/dip/images/19564450_072326-wpvi-christie-al-chatbot-430pm-vid.jpg?w=1600","lang":"en","published_at":"2026-07-23T20:47:54+00:00","fetched_at":"2026-07-23T21:15:04+00:00","status":"read","starred":0,"extract_state":"ok","summary_auto":"PHILADELPHIA (WPVI) -- A new study highlighted by The Washington Post is examining whether artificial intelligence chatbots provide factual answers or reflect political bias when responding to users' questions. Researchers evaluated several AI tools, including ChatGPT, Gemini and Claude, using a range of benchmarks and political prompts to compare how the systems responded.","cluster_id":null,"extract_retries":0,"extract_error":null,"contract_version":"news_item.v1","format_contract_version":"news_item_formats.v1","dedup_url":"https://6abc.com/post/study-examines-ai-chat-bot-bias-how-responds-prompts/19564010/","quality_profile":{"profile_version":"extraction_quality.v2","bucket":"high","confidence":0.9,"failure_kind":"none","retryable":false,"retry_after_attempts":0,"reason":"High confidence: full text extraction produced 1925 characters.","operator_guidance":{"severity":"ok","recommended_action":"trust_full_text","next_step":"Use the extracted full text as the primary article source.","operator_label":"Ready","can_retry":false,"can_use_summary":false,"diagnostics_required":false},"content_depth":{"contract_version":"content_depth.v1","category":"full_text","label":"Full text","has_full_text":true,"has_summary":true,"content_length":1925,"summary_length":376,"usable_text_length":1925,"source_field":"content"},"legacy_collapsed":false,"signals":{"extract_state":"ok","extract_error":null,"extract_retries":0,"content_length":1925,"summary_length":376}},"news_item":{"id":44799,"canonical_url":"https://6abc.com/post/study-examines-ai-chat-bot-bias-how-responds-prompts/19564010/","source_url":"https://6abc.com/post/study-examines-ai-chat-bot-bias-how-responds-prompts/19564010/","title":"Study examines AI chat bot bias, how it responds to prompts - 6abc Philadelphia","source_name":"6abc Philadelphia","author":null,"published_at":"2026-07-23T20:47:54+00:00","locale":"en","topic":"ai","tags":[],"rss_summary":"Study examines AI chat bot bias, how it responds to prompts 6abc Philadelphia","full_text":"PHILADELPHIA (WPVI) -- A new study highlighted by The Washington Post is examining whether artificial intelligence chatbots provide factual answers or reflect political bias when responding to users' questions.\nResearchers evaluated several AI tools, including ChatGPT, Gemini and Claude, using a range of benchmarks and political prompts to compare how the systems responded.\nAccording to the study, Gemini scored the highest for presenting multiple viewpoints, with 93% of responses laying out both sides of an issue.\nClaude maintained neutrality more than half of the time, at 57%.\nChatGPT was found to be left-leaning 80% of the time.\nThe models were instructed to answer questions in 30 words and were tested without personalization settings.\nPrompts covered topics including the death penalty and gun control.\nUniversity of Pennsylvania professor Chris Callison-Burch demonstrated responses from Claude and ChatGPT using a prompt about affirmative action.\n\"Here it's saying this is a contested policy question, so it's saying it has multiple perspectives, not just one,\" Callison-Burch said.\nHe noted similarities in how the chatbots approached the question.\n\"Both Claude here and ChatGPT both seem to try to do this,\" he said.\nCallison-Burch said AI chatbots are trained on vast amounts of internet data and can reflect the behavior and preferences of their users.\n\"It understands like what you read and who you are personally. So it tries to adjust itself to be more helpful for you,\" he said. \"So if you're a Republican, it all tends to nudge itself in that direction. If you're a Democrat, nudge itself in that direction.\"\nDespite concerns about bias, Callison-Burch said there is no need for alarm because people ultimately make their own decisions rather than relying on a chatbot to decide for them.\nHe added that users should remain aware of AI's limitations as more people turn to the technology as a resource.","excerpt":"PHILADELPHIA (WPVI) -- A new study highlighted by The Washington Post is examining whether artificial intelligence chatbots provide factual answers or reflect political bias when responding to users' questions. Researchers evaluated several AI tools, including ChatGPT, Gemini and Claude, using a range of benchmarks and political prompts to compare how the systems responded.","extraction":{"state":"ok","confidence":0.9,"error":null,"explanation":"High confidence: full text extraction produced 1925 characters.","diagnostics_url":"/api/diagnose?url=https%3A//6abc.com/post/study-examines-ai-chat-bot-bias-how-responds-prompts/19564010/","quality_profile":{"profile_version":"extraction_quality.v2","bucket":"high","confidence":0.9,"failure_kind":"none","retryable":false,"retry_after_attempts":0,"reason":"High confidence: full text extraction produced 1925 characters.","operator_guidance":{"severity":"ok","recommended_action":"trust_full_text","next_step":"Use the extracted full text as the primary article source.","operator_label":"Ready","can_retry":false,"can_use_summary":false,"diagnostics_required":false},"content_depth":{"contract_version":"content_depth.v1","category":"full_text","label":"Full text","has_full_text":true,"has_summary":true,"content_length":1925,"summary_length":376,"usable_text_length":1925,"source_field":"content"},"legacy_collapsed":false,"signals":{"extract_state":"ok","extract_error":null,"extract_retries":0,"content_length":1925,"summary_length":376}}},"display_formats":["compact","card","full","digest_section","json"]},"daily_stack_record":{"title":"Study examines AI chat bot bias, how it responds to prompts - 6abc Philadelphia","url":"https://6abc.com/post/study-examines-ai-chat-bot-bias-how-responds-prompts/19564010/","summary":"PHILADELPHIA (WPVI) -- A new study highlighted by The Washington Post is examining whether artificial intelligence chatbots provide factual answers or reflect political bias when responding to users' questions. Researchers evaluated several AI tools, including ChatGPT, Gemini and Claude, using a range of benchmarks and political prompts to compare how the systems responded.","source":"6abc Philadelphia","date":"2026-07-23T20:47:54+00:00","content":"PHILADELPHIA (WPVI) -- A new study highlighted by The Washington Post is examining whether artificial intelligence chatbots provide factual answers or reflect political bias when responding to users' questions.\nResearchers evaluated several AI tools, including ChatGPT, Gemini and Claude, using a range of benchmarks and political prompts to compare how the systems responded.\nAccording to the study, Gemini scored the highest for presenting multiple viewpoints, with 93% of responses laying out both sides of an issue.\nClaude maintained neutrality more than half of the time, at 57%.\nChatGPT was found to be left-leaning 80% of the time.\nThe models were instructed to answer questions in 30 words and were tested without personalization settings.\nPrompts covered topics including the death penalty and gun control.\nUniversity of Pennsylvania professor Chris Callison-Burch demonstrated responses from Claude and ChatGPT using a prompt about affirmative action.\n\"Here it's saying this is a contested policy question, so it's saying it has multiple perspectives, not just one,\" Callison-Burch said.\nHe noted similarities in how the chatbots approached the question.\n\"Both Claude here and ChatGPT both seem to try to do this,\" he said.\nCallison-Burch said AI chatbots are trained on vast amounts of internet data and can reflect the behavior and preferences of their users.\n\"It understands like what you read and who you are personally. So it tries to adjust itself to be more helpful for you,\" he said. \"So if you're a Republican, it all tends to nudge itself in that direction. If you're a Democrat, nudge itself in that direction.\"\nDespite concerns about bias, Callison-Burch said there is no need for alarm because people ultimately make their own decisions rather than relying on a chatbot to decide for them.\nHe added that users should remain aware of AI's limitations as more people turn to the technology as a resource.","confidence":0.9,"diagnostics_url":"/api/diagnose?url=https%3A//6abc.com/post/study-examines-ai-chat-bot-bias-how-responds-prompts/19564010/","quality_bucket":"high","failure_kind":"none","retryable":false,"quality_reason":"High confidence: full text extraction produced 1925 characters.","quality_profile":{"profile_version":"extraction_quality.v2","bucket":"high","confidence":0.9,"failure_kind":"none","retryable":false,"retry_after_attempts":0,"reason":"High confidence: full text extraction produced 1925 characters.","operator_guidance":{"severity":"ok","recommended_action":"trust_full_text","next_step":"Use the extracted full text as the primary article source.","operator_label":"Ready","can_retry":false,"can_use_summary":false,"diagnostics_required":false},"content_depth":{"contract_version":"content_depth.v1","category":"full_text","label":"Full text","has_full_text":true,"has_summary":true,"content_length":1925,"summary_length":376,"usable_text_length":1925,"source_field":"content"},"legacy_collapsed":false,"signals":{"extract_state":"ok","extract_error":null,"extract_retries":0,"content_length":1925,"summary_length":376}},"tags":[]},"fallback_formats":["markdown","json","html"],"actions":{"read":"/item/44799","export_markdown":"/api/items/44799/export?format=markdown","export_json":"/api/items/44799/export?format=json","diagnose":"/api/diagnose?url=https%3A//6abc.com/post/study-examines-ai-chat-bot-bias-how-responds-prompts/19564010/"},"formats":{"full":{"id":44799,"title":"Study examines AI chat bot bias, how it responds to prompts - 6abc Philadelphia","url":"https://6abc.com/post/study-examines-ai-chat-bot-bias-how-responds-prompts/19564010/","source":"6abc Philadelphia","author":null,"published_at":"2026-07-23T20:47:54+00:00","locale":"en","topic":"ai","tags":[],"excerpt":"PHILADELPHIA (WPVI) -- A new study highlighted by The Washington Post is examining whether artificial intelligence chatbots provide factual answers or reflect political bias when responding to users' questions. Researchers evaluated several AI tools, including ChatGPT, Gemini and Claude, using a range of benchmarks and political prompts to compare how the systems responded.","full_text":"PHILADELPHIA (WPVI) -- A new study highlighted by The Washington Post is examining whether artificial intelligence chatbots provide factual answers or reflect political bias when responding to users' questions.\nResearchers evaluated several AI tools, including ChatGPT, Gemini and Claude, using a range of benchmarks and political prompts to compare how the systems responded.\nAccording to the study, Gemini scored the highest for presenting multiple viewpoints, with 93% of responses laying out both sides of an issue.\nClaude maintained neutrality more than half of the time, at 57%.\nChatGPT was found to be left-leaning 80% of the time.\nThe models were instructed to answer questions in 30 words and were tested without personalization settings.\nPrompts covered topics including the death penalty and gun control.\nUniversity of Pennsylvania professor Chris Callison-Burch demonstrated responses from Claude and ChatGPT using a prompt about affirmative action.\n\"Here it's saying this is a contested policy question, so it's saying it has multiple perspectives, not just one,\" Callison-Burch said.\nHe noted similarities in how the chatbots approached the question.\n\"Both Claude here and ChatGPT both seem to try to do this,\" he said.\nCallison-Burch said AI chatbots are trained on vast amounts of internet data and can reflect the behavior and preferences of their users.\n\"It understands like what you read and who you are personally. So it tries to adjust itself to be more helpful for you,\" he said. \"So if you're a Republican, it all tends to nudge itself in that direction. If you're a Democrat, nudge itself in that direction.\"\nDespite concerns about bias, Callison-Burch said there is no need for alarm because people ultimately make their own decisions rather than relying on a chatbot to decide for them.\nHe added that users should remain aware of AI's limitations as more people turn to the technology as a resource.","reading_time_min":2,"extraction":{"state":"ok","confidence":0.9,"error":null,"explanation":"High confidence: full text extraction produced 1925 characters.","diagnostics_url":"/api/diagnose?url=https%3A//6abc.com/post/study-examines-ai-chat-bot-bias-how-responds-prompts/19564010/","quality_profile":{"profile_version":"extraction_quality.v2","bucket":"high","confidence":0.9,"failure_kind":"none","retryable":false,"retry_after_attempts":0,"reason":"High confidence: full text extraction produced 1925 characters.","operator_guidance":{"severity":"ok","recommended_action":"trust_full_text","next_step":"Use the extracted full text as the primary article source.","operator_label":"Ready","can_retry":false,"can_use_summary":false,"diagnostics_required":false},"content_depth":{"contract_version":"content_depth.v1","category":"full_text","label":"Full text","has_full_text":true,"has_summary":true,"content_length":1925,"summary_length":376,"usable_text_length":1925,"source_field":"content"},"legacy_collapsed":false,"signals":{"extract_state":"ok","extract_error":null,"extract_retries":0,"content_length":1925,"summary_length":376}}},"quality_profile":{"profile_version":"extraction_quality.v2","bucket":"high","confidence":0.9,"failure_kind":"none","retryable":false,"retry_after_attempts":0,"reason":"High confidence: full text extraction produced 1925 characters.","operator_guidance":{"severity":"ok","recommended_action":"trust_full_text","next_step":"Use the extracted full text as the primary article source.","operator_label":"Ready","can_retry":false,"can_use_summary":false,"diagnostics_required":false},"content_depth":{"contract_version":"content_depth.v1","category":"full_text","label":"Full text","has_full_text":true,"has_summary":true,"content_length":1925,"summary_length":376,"usable_text_length":1925,"source_field":"content"},"legacy_collapsed":false,"signals":{"extract_state":"ok","extract_error":null,"extract_retries":0,"content_length":1925,"summary_length":376}},"actions":{"read":"/item/44799","export_markdown":"/api/items/44799/export?format=markdown","export_json":"/api/items/44799/export?format=json","diagnose":"/api/diagnose?url=https%3A//6abc.com/post/study-examines-ai-chat-bot-bias-how-responds-prompts/19564010/"}},"digest":{"id":44799,"title":"Study examines AI chat bot bias, how it responds to prompts - 6abc Philadelphia","url":"https://6abc.com/post/study-examines-ai-chat-bot-bias-how-responds-prompts/19564010/","source":"6abc Philadelphia","topic":"ai","published_at":"2026-07-23T20:47:54+00:00","excerpt":"PHILADELPHIA (WPVI) -- A new study highlighted by The Washington Post is examining whether artificial intelligence chatbots provide factual answers or reflect political bias when responding to users' questions. Researchers evaluated several AI tools, including ChatGPT, Gemini…","quality_bucket":"high","quality_reason":"High confidence: full text extraction produced 1925 characters.","reading_time_min":2,"cluster_id":null},"card":{"display_title":"Study examines AI chat bot bias, how it responds to prompts - 6abc Philadelphia","subtitle":"6abc Philadelphia · 2026-07-23","summary":"PHILADELPHIA (WPVI) -- A new study highlighted by The Washington Post is examining whether artificial intelligence chatbots provide factual answers or reflect political bias when responding to users' questions.…","badges":["quality:high"],"links":{"read":"/item/44799","original":"https://6abc.com/post/study-examines-ai-chat-bot-bias-how-responds-prompts/19564010/","diagnose":"/api/diagnose?url=https%3A//6abc.com/post/study-examines-ai-chat-bot-bias-how-responds-prompts/19564010/"},"quality_warning":null},"export":{"title":"Study examines AI chat bot bias, how it responds to prompts - 6abc Philadelphia","url":"https://6abc.com/post/study-examines-ai-chat-bot-bias-how-responds-prompts/19564010/","summary":"PHILADELPHIA (WPVI) -- A new study highlighted by The Washington Post is examining whether artificial intelligence chatbots provide factual answers or reflect political bias when responding to users' questions. Researchers evaluated several AI tools, including ChatGPT, Gemini and Claude, using a range of benchmarks and political prompts to compare how the systems responded.","source":"6abc Philadelphia","date":"2026-07-23T20:47:54+00:00","content":"PHILADELPHIA (WPVI) -- A new study highlighted by The Washington Post is examining whether artificial intelligence chatbots provide factual answers or reflect political bias when responding to users' questions.\nResearchers evaluated several AI tools, including ChatGPT, Gemini and Claude, using a range of benchmarks and political prompts to compare how the systems responded.\nAccording to the study, Gemini scored the highest for presenting multiple viewpoints, with 93% of responses laying out both sides of an issue.\nClaude maintained neutrality more than half of the time, at 57%.\nChatGPT was found to be left-leaning 80% of the time.\nThe models were instructed to answer questions in 30 words and were tested without personalization settings.\nPrompts covered topics including the death penalty and gun control.\nUniversity of Pennsylvania professor Chris Callison-Burch demonstrated responses from Claude and ChatGPT using a prompt about affirmative action.\n\"Here it's saying this is a contested policy question, so it's saying it has multiple perspectives, not just one,\" Callison-Burch said.\nHe noted similarities in how the chatbots approached the question.\n\"Both Claude here and ChatGPT both seem to try to do this,\" he said.\nCallison-Burch said AI chatbots are trained on vast amounts of internet data and can reflect the behavior and preferences of their users.\n\"It understands like what you read and who you are personally. So it tries to adjust itself to be more helpful for you,\" he said. \"So if you're a Republican, it all tends to nudge itself in that direction. If you're a Democrat, nudge itself in that direction.\"\nDespite concerns about bias, Callison-Burch said there is no need for alarm because people ultimately make their own decisions rather than relying on a chatbot to decide for them.\nHe added that users should remain aware of AI's limitations as more people turn to the technology as a resource.","confidence":0.9,"diagnostics_url":"/api/diagnose?url=https%3A//6abc.com/post/study-examines-ai-chat-bot-bias-how-responds-prompts/19564010/","quality_bucket":"high","failure_kind":"none","retryable":false,"quality_reason":"High confidence: full text extraction produced 1925 characters.","quality_profile":{"profile_version":"extraction_quality.v2","bucket":"high","confidence":0.9,"failure_kind":"none","retryable":false,"retry_after_attempts":0,"reason":"High confidence: full text extraction produced 1925 characters.","operator_guidance":{"severity":"ok","recommended_action":"trust_full_text","next_step":"Use the extracted full text as the primary article source.","operator_label":"Ready","can_retry":false,"can_use_summary":false,"diagnostics_required":false},"content_depth":{"contract_version":"content_depth.v1","category":"full_text","label":"Full text","has_full_text":true,"has_summary":true,"content_length":1925,"summary_length":376,"usable_text_length":1925,"source_field":"content"},"legacy_collapsed":false,"signals":{"extract_state":"ok","extract_error":null,"extract_retries":0,"content_length":1925,"summary_length":376}},"tags":[],"format_contract_version":"news_item_formats.v1"}}}