{
  "$type": "site.standard.document",
  "bskyPostRef": {
    "cid": "bafyreibehmmixnhgpc4qokvg2icgeb73wy6ugu25vhg7gtxt5lctna55za",
    "uri": "at://did:plc:qifj232pstd6e5342l763se5/app.bsky.feed.post/3mjunppym7si2"
  },
  "path": "/news/112117-data-failed-startups-finds-second-life-ai-training.html",
  "publishedAt": "2026-04-19T16:19:00.000Z",
  "site": "https://www.techspot.com",
  "tags": [
    "Read Entire Article"
  ],
  "textContent": "\nWhat was once considered operational residue is now being packaged, scrubbed, and sold to AI developers seeking richer training environments. The shift reflects a broader evolution in how advanced AI models are built. Early large language models drew heavily from news archives, Wikipedia, and forums. Now, newer systems, particularly agentic...\n\nRead Entire Article\n\n",
  "title": "Slack chats and internal data from failed startups are finding a second life in AI training"
}