{"685420":{"#nid":"685420","#data":{"type":"external_news","title":"Will we know artificial general intelligence when we see it?","body":[{"value":"\u003Cp dir=\u0022ltr\u0022\u003EWe may never agree on what AGI or \u201chumanlike\u201d AI means, or what suffices to prove it. As AI advances, machines will still make mistakes, and people will point to these and say the AIs aren\u2019t really intelligent.\u0026nbsp;\u003Ca href=\u0022https:\/\/psychology.gatech.edu\/people\/anna-ivanova\u0022\u003E\u003Cstrong\u003EAnna Ivanova\u003C\/strong\u003E\u003C\/a\u003E, an assistant professor in the\u0026nbsp;\u003Ca href=\u0022https:\/\/psychology.gatech.edu\/\u0022\u003ESchool of Psychology\u003C\/a\u003E at Georgia Tech, was on a panel recently, and the moderator asked about AGI timelines. \u201cWe had one person saying that it might never happen,\u201d Ivanova told me, \u201cand one person saying that it already happened.\u201d So the term \u201cAGI\u201d may be convenient shorthand to express an aim\u2014or a fear\u2014but its practical use may be limited. In most cases, it should come with an asterisk, and a benchmark.\u003C\/p\u003E","summary":"","format":"limited_html"}],"field_subtitle":"","field_summary":[{"value":"\u003Cp dir=\u0022ltr\u0022\u003EWe may never agree on what AGI or \u201chumanlike\u201d AI means, or what suffices to prove it. As AI advances, machines will still make mistakes, and people will point to these and say the AIs aren\u2019t really intelligent.\u0026nbsp;\u003Ca href=\u0022https:\/\/psychology.gatech.edu\/people\/anna-ivanova\u0022\u003E\u003Cstrong\u003EAnna Ivanova\u003C\/strong\u003E\u003C\/a\u003E, an assistant professor in the\u0026nbsp;\u003Ca href=\u0022https:\/\/psychology.gatech.edu\/\u0022\u003ESchool of Psychology\u003C\/a\u003E at Georgia Tech, was on a panel recently, and the moderator asked about AGI timelines. \u201cWe had one person saying that it might never happen,\u201d Ivanova told me, \u201cand one person saying that it already happened.\u201d So the term \u201cAGI\u201d may be convenient shorthand to express an aim\u2014or a fear\u2014but its practical use may be limited. In most cases, it should come with an asterisk, and a benchmark.\u003C\/p\u003E","format":"limited_html"}],"field_summary_sentence":"","uid":"36583","created_gmt":"2025-09-30 21:40:33","changed_gmt":"2025-10-01 19:57:50","author":"lvidal7","boilerplate_text":"","field_publication":"","publication":"IEEE Spectrum","field_article_url":"","publication_url":"https:\/\/spectrum.ieee.org\/agi-benchmark","dateline":{"date":"2025-09-22T00:00:00-04:00","iso_date":"2025-09-22T00:00:00-04:00","tz":"America\/New_York"},"extras":[],"groups":[{"id":"1278","name":"College of Sciences"},{"id":"443951","name":"School of Psychology"}],"categories":[{"id":"194606","name":"Artificial Intelligence"},{"id":"135","name":"Research"},{"id":"134","name":"Student and Faculty"}],"keywords":[{"id":"192258","name":"cos-data"},{"id":"4896","name":"College of Sciences"},{"id":"167710","name":"School of Psychology"},{"id":"2556","name":"artificial intelligence"}],"core_research_areas":[],"news_room_topics":[],"event_categories":[],"invited_audience":[],"affiliations":[],"classification":[],"areas_of_expertise":[],"news_and_recent_appearances":[],"phone":[],"contact":[],"email":[],"slides":[],"orientation":[],"userdata":""}}}