{
  "$type": "site.standard.document",
  "bskyPostRef": {
    "cid": "bafyreif6ygp4oqxoc7th4la6ztecpftqip2s7xotqorzwktnk3sjybes2y",
    "uri": "at://did:plc:s6aepfdltjflp62edxfnv35b/app.bsky.feed.post/3mj54mbhpma32"
  },
  "path": "/2026/04/09/researchers-detail-how-a-prompt-injection-attack-bypassed-apple-intelligence-protections/",
  "publishedAt": "2026-04-09T20:39:06.000Z",
  "site": "https://9to5mac.com",
  "tags": [
    "News",
    "more…"
  ],
  "textContent": "A now corrected issue allowed researchers to circumvent Apple’s restrictions and force the on-device LLM to execute attacker-controlled actions. Here’s how they did it.\n\nmore…",
  "title": "Researchers detail how a prompt injection attack bypassed Apple Intelligence protections"
}