{
  "$type": "site.standard.document",
  "bskyPostRef": {
    "cid": "bafyreicom46jmcyh37imjbjlzwm4skuwaxoooaq3gc7vb2zelzxmugawde",
    "uri": "at://did:plc:lk3jfj3zq4k4wxnk474axylu/app.bsky.feed.post/3mnozskon3lk2"
  },
  "path": "/t/how-to-stop-codex-from-rushing-fixes/1382830?page=2#post_27",
  "publishedAt": "2026-06-07T10:38:58.000Z",
  "site": "https://community.openai.com",
  "textContent": "shamimarakhaliq:\n\n> I think any AI understands these words better than me.\n\nThe issue is that they do not understand words… (“understand” I define as successful decoding of the meaning originally put into the information).\n\nThey perceive ALL the meanings possibly encoded in the information and select the meaning that is most likely to fit based on the surrounding patterns of the language present in the current context.\n\nYou’re right about the biases, but their bias is way more complicated than what we can handle with the precision we want.\n\nYou define the words (especially common, overloaded with meanings) NOT FOR THEM TO UNDERSTAND THE WORD, BUT TO DEFINE WHAT YOU MEAN WHEN YOU USE THAT EXACT WORD.\n\nBecause you would not keep up with them to analyze the patterns in your own message and make sure your meaning is transmitted the way you need.\n\nThe samples you gave me, on my opinion are the exact way to get low quality results and hallucinations, because they lack clarity.",
  "title": "How to stop Codex from rushing fixes?"
}