{
  "$type": "site.standard.document",
  "bskyPostRef": {
    "cid": "bafyreiftnwyxv3hc3ukorgutdq5jkze2w2v3vwqksn5rnw46ne64juohci",
    "uri": "at://did:plc:lk3jfj3zq4k4wxnk474axylu/app.bsky.feed.post/3mmld434q2vp2"
  },
  "path": "/t/openai-is-making-a-huge-mistake-by-deprecating-dall-e-3/1367228?page=5#post_86",
  "publishedAt": "2026-05-24T04:29:12.000Z",
  "site": "https://community.openai.com",
  "textContent": "windysoliloquy:\n\n> You guys know what I’m thinking of?\n\nWhat gpt-image-2 and multimodal learners can’t be: an unimaginable spectrum of keyword hallucinations evolved out of nothingness.\n\nTiny but coherent (well..sort of).\n\nWhere it also can’t succeed currently is in the artifact-free reality detailed deep in every pixel that others might provide.",
  "title": "OpenAI is making a huge mistake by deprecating DALL-E-3"
}