{
"$type": "site.standard.document",
"bskyPostRef": {
"cid": "bafyreifbhl4cbuvfnlijxx2om43gumbcawbykx7s52eu7o7gx55nvqcqfy",
"uri": "at://did:plc:iir655mcoipvnewhnkv6fb3u/app.bsky.feed.post/3mnrrqjgwn4f2"
},
"coverImage": {
"$type": "blob",
"ref": {
"$link": "bafkreiajapwa4ilec2df2rkwpqcw42u4b2fteroh7sii6e27tfun7aujmm"
},
"mimeType": "image/jpeg",
"size": 52814
},
"path": "/lockdown-mode-means-default-chatgpt-wasnt-safe/",
"publishedAt": "2026-06-08T10:27:57.000Z",
"site": "https://thecyberexpress.com",
"textContent": "\nOpenAI introduced two new protections designed to help users and organizations mitigate prompt injection attacks when it launched Lockdown Mode in February.",
"title": "Is OpenAI’s New Lockdown Mode an Admission That Default ChatGPT Was Never Safe Enough?"
}