{
"$type": "site.standard.document",
"bskyPostRef": {
"cid": "bafyreihr3bxhjwbflbcka7fd22vsyobyfk3ditnt6h75zkp6emmyl5hk4y",
"uri": "at://did:plc:iir655mcoipvnewhnkv6fb3u/app.bsky.feed.post/3mf4yqrsnlic2"
},
"coverImage": {
"$type": "blob",
"ref": {
"$link": "bafkreidudhs2qfv7tfvhwn2ytozv3ehyl7juj35oechcqkur3qlibb4uga"
},
"mimeType": "image/webp",
"size": 92650
},
"path": "/openai-new-lockdown-mode/",
"publishedAt": "2026-02-18T10:05:06.000Z",
"site": "https://thecyberexpress.com",
"tags": [
"link"
],
"textContent": "\nOpenAI deployed two security features, Lockdown Mode and Elevated Risk labels, targeting prompt injection attacks that exploit AI systems' growing connectivity to external networks and applications.",
"title": "OpenAI’s New Enterprise Security Mode Locks Down ChatGPT Against Prompt Injection"
}