{
"$type": "site.standard.document",
"bskyPostRef": {
"cid": "bafyreifcsnmqku5qw73md2tmwktzbkqmydow3veyebj23uw7jq2fifl23a",
"uri": "at://did:plc:lk3jfj3zq4k4wxnk474axylu/app.bsky.feed.post/3mlsk6o2m2go2"
},
"path": "/t/experimenting-with-codex-deciding-its-own-next-steps/1380898#post_4",
"publishedAt": "2026-05-14T09:49:16.000Z",
"site": "https://community.openai.com",
"textContent": "Hi and welcome to the community!\n\nVery interesting!\n\nWhat is your experience with models choosing shortcuts to reach the goal instead of actually doing the work?\n\nI ran into this a few times recently. I was too optimistic and expected the model to follow through, but instead it found the shortest path to satisfy poorly defined goal criteria.\n\nNeedless to say, I had to go back to the basics and implement a proper sequence of gates before continuing with this approach.",
"title": "Experimenting with Codex deciding its own next steps"
}