Dragon-sided D on Nostr: "[T]he jailbreak was shown to work on [...] * Meta Llama3-70b-instruct * Google ...
Published at
2024-06-27 23:04:55Event JSON
{
"id": "07c0a073ff5be3f4ba12d35bb0f30703d962929d979f38ff8d9b443d14b531fc",
"pubkey": "867b16fbdb961d3620b6e6c1dd0b0124703c4db31d6d71096b866e54dec93560",
"created_at": 1719529495,
"kind": 1,
"tags": [
[
"proxy",
"https://sciencemastodon.com/users/dragonsidedd/statuses/112691085001602786",
"activitypub"
]
],
"content": "\"[T]he jailbreak was shown to work on [...]\n* Meta Llama3-70b-instruct\n* Google Gemini Pro\n* OpenAI GPT 3.5 Turbo\n* OpenAI GPT 4o\n* Mistral Large\n* Anthropic Claude 3 Opus\n* Cohere Commander R Plus\n\"\n\nhttps://www.microsoft.com/en-us/security/blog/2024/06/26/mitigating-skeleton-key-a-new-type-of-generative-ai-jailbreak-technique/\n\nhttps://cdn.masto.host/sciencemastodoncom/media_attachments/files/112/691/066/325/626/801/original/5c55a19a39482239.png",
"sig": "3ffc5496fcae2200aea4ba39aadba93a39152791bbba277458f9f195c16a943c8fa0d71940d26e14426eb7a6137d1172bea5b0d07535630b4f1023ea76dd0dd1"
}