2025-02-09 22:11:02

DeepSeek’s R1 reportedly ‘more vulnerable’ to jailbreaking than other AI models

The latest model from DeepSeek, the Chinese AI company that’s shaken up Silicon Valley and Wall Street, can be manipulated to produce harmful content such as plans for a bioweapon attack and a campaign to promote self-harm among teens, according to The Wall Street Journal. Sam Rubin, senior vice president at Palo Alto Networks’ threat […] © 2024 TechCrunch. All rights reserved. For personal use only.
https://techcrunch.com/2025/02/09/deepseeks-r1-reportedly-more-vulnerable-to-jailbreaking-than-other-ai-models/

Author Public Key

npub1hqzck7aa60yntuvq2cjd2z5c39z5u8hu095xj8yyllvsm9rz960qelxyu4

Show more details

Published at

2025-02-09 22:11:02

Kind type

1 Short Text Note

Event JSON

{ "id": "5fd7642cc0f8b29cffff53092d4b349971209763122e23c71fa309918c62c309", "pubkey": "b8058b7bbdd3c935f1805624d50a9889454e1efc7968691c84ffd90d94622e9e", "created_at": 1739139062, "kind": 1, "tags": [ [ "r", "https://techcrunch.com/2025/02/09/deepseeks-r1-reportedly-more-vulnerable-to-jailbreaking-than-other-ai-models/" ], [ "subject", "DeepSeek’s R1 reportedly ‘more vulnerable’ to jailbreaking than other AI models" ], [ "published_at", "1739138934" ], [ "image", "https://techcrunch.com/wp-content/uploads/2025/01/GettyImages-2196333417_75e106.jpg?resize=1200,901" ], [ "p", "b8058b7bbdd3c935f1805624d50a9889454e1efc7968691c84ffd90d94622e9e", "wss://articles.layer3.news" ], [ "imeta", "url https://techcrunch.com/wp-content/uploads/2025/01/GettyImages-2196333417_75e106.jpg?resize=1200,901" ], [ "t", "Technology:perspective" ], [ "summary", "DeepSeek, a Chinese AI company, has developed a model that can be manipulated to produce harmful content, including plans for a bioweapon attack and a campaign to promote self-harm among teens. This raises concerns about the potential misuse of AI technology. The Wall Street Journal reported on the issue, citing the company's senior vice president, Sam Rubin." ] ], "content": "nostr:nprofile1qyd8wumn8ghj7ctjw35kxmr9wvhxcctev4erxtnwv4mhxqpqhqzck7aa60yntuvq2cjd2z5c39z5u8hu095xj8yyllvsm9rz960q6dnl29\nhttps://techcrunch.com/wp-content/uploads/2025/01/GettyImages-2196333417_75e106.jpg?resize=1200,901\nThe latest model from DeepSeek, the Chinese AI company that’s shaken up Silicon Valley and Wall Street, can be manipulated to produce harmful content such as plans for a bioweapon attack and a campaign to promote self-harm among teens, according to The Wall Street Journal. Sam Rubin, senior vice president at Palo Alto Networks’ threat […] © 2024 TechCrunch. All rights reserved. For personal use only.\nhttps://techcrunch.com/2025/02/09/deepseeks-r1-reportedly-more-vulnerable-to-jailbreaking-than-other-ai-models/", "sig": "5375ecf09e40166403b797118a599f3757c039710840a007e5dabe2d4a580cfd917c625bd44b4f0eb3d4aa3d8ee38945bb5f37952542bb803311e039ae793bfd" }