{
"$type": "site.standard.document",
"bskyPostRef": {
"cid": "bafyreibovcf5lo2oszrwq45xcmwiv46yqfcjmjloe53fhb5o2jsg7fyxpy",
"uri": "at://did:plc:ctcqaqa3tlbhfwmfefti346h/app.bsky.feed.post/3mejxrp27mor2"
},
"coverImage": {
"$type": "blob",
"ref": {
"$link": "bafkreibwtkgjmlwanlf7af42rryziur2pimqf5yjmkqamx2ezmqhugk3we"
},
"mimeType": "image/jpeg",
"size": 144479
},
"path": "/pro/as-adoption-grows-confidence-in-safeguards-must-rise-with-it-microsoft-reveals-new-tool-which-can-track-backdoors-in-llms-and-its-hoping-this-will-restore-trust-in-ai-across-the-world",
"publishedAt": "2026-02-10T20:20:00.000Z",
"site": "https://www.techradar.com",
"tags": [
"Security",
"Pro"
],
"textContent": "Microsoft introduced a scanner that detects poisoned open-weight language models by analyzing attention behavior, memorization leaks, and trigger flexibility.",
"title": "'As adoption grows, confidence in safeguards must rise with it': Microsoft reveals new tool which can track backdoors in LLMs - and it's hoping this will restore trust in AI across the world",
"updatedAt": "2026-02-10T16:22:02.000Z"
}