{"id":"2073103304612012202","url":"https://x.com/poetengineer__/status/2073103304612012202","text":"what does ai actually add to open-ended creative work? my first attempt at giving it a process signature: from a real project's transcripts i pulled 146 ideation events; the cubes are the moments ai forked the ideas open, laid along a 3d chronological trace of the sessions.","author":{"name":"Kat ⊷ the Poet Engineer","username":"poetengineer__","avatarUrl":"https://pbs.twimg.com/profile_images/1783930619782656000/3hXoqxDX_200x200.jpg"},"createdAt":"Fri Jul 03 17:55:18 +0000 2026","engagement":{"replies":31,"retweets":77,"likes":1042,"views":68671},"media":{"photos":[],"videos":[{"url":"https://video.twimg.com/amplify_video/2073102931105132544/vid/avc1/1886x1080/mi6FRZ6nawfWVfZ9.mp4?tag=28","thumbnailUrl":"https://pbs.twimg.com/amplify_video_thumb/2073102931105132544/img/akjJvH-i_2umSuzG.jpg","width":1886,"height":1080,"duration":26.71}]},"quoteTweet":{"id":"2061546705217057101","url":"https://x.com/poetengineer__/status/2061546705217057101","text":"things i find severely lacking in current llm benchmarks but i feel are important for iterative, open-ended knowledge work:\n- problem-setting. a major part of creative/exploratory work is defining what the problem even looks like. benchmarks skip that and hand you the problem pre-formed. it's the same issue with using exams to predict how someone handles real world problems.","author":{"name":"Kat ⊷ the Poet Engineer","username":"poetengineer__","avatarUrl":"https://pbs.twimg.com/profile_images/1783930619782656000/3hXoqxDX_200x200.jpg"},"createdAt":"Mon Jun 01 20:33:30 +0000 2026"},"adhxContext":{"savedByCount":1,"publicTags":[],"previewUrl":"https://adhx.com/poetengineer__/status/2073103304612012202"}}