Aggregator

随着大规模生成式模型在各行业中的应用，模型的安全性问题逐渐暴露，尤其是提示词泄漏问题。提示词泄漏是攻击者通过精心设计的提示词，诱导模型暴露内部敏感数据，造成企业或个人隐私数据被侵犯，甚至可能对政治倾向等方面产生影响。

随着大规模生成式模型在各行业中的应用，模型的安全性问题逐渐暴露，尤其是提示词泄漏问题。提示词泄漏是攻击者通过精心设计的提示词，诱导模型暴露内部敏感数据，造成企业或个人隐私数据被侵犯，甚至可能对政治倾向等方面产生影响。

背景随着大规模生成式模型在各行业中的应用，模型的安全性问题逐渐暴露，尤其是提示词泄漏问题。提示词泄漏是攻击者通过精心设计的提示词，诱导模型暴露内部敏感数据，造成企业或个人隐私数据被侵犯，甚至可能对政治

You should think twice before trusting your AI assistant, as database poisoning can markedly alter its output – even dangerously so

The UK’s National Cyber Security Centre has released a new paper making it easier to assess if a flaw is “unforgivable”

One day, you wake up and decide you can't live without TypeScript in your legacy JavaScript project

In recent years, the healthcare sector has emerged as a primary target for cyberattacks, which is mainly due to the highly sensitive nature of medical information.

The post Transforming Healthcare Security: Why Zero-Trust is Essential appeared first on Security Boulevard.

The COVID-19 pandemic accelerated the digital transformation of healthcare services, leading to

攻击指令激增上百倍

关注我们带你读懂网络安全奇安信安全专家：针对DeepSeek的网络攻击再升级，攻击指令激增上百倍。1月30日凌晨，即农历大年初二，奇安信XLab实验室监测发现，针对DeepSeek（深度求索）线上服务

We’ve developed a new database design tool called Liam ERD, and it’s finally here! Let us introduce

Фальшивые резюме, ядерные секреты… На что способен ИИ под руководством спецслужб?

A vulnerability was found in Apple macOS up to 13.2.1. It has been declared as problematic. This vulnerability affects unknown code of the component TCC. The manipulation leads to improper access controls. This vulnerability was named CVE-2023-27931. The attack needs to be approached locally. There is no exploit available. It is recommended to upgrade the affected component.

A vulnerability was found in Apple iOS and iPadOS up to 16.3.1. It has been rated as problematic. This issue affects some unknown processing of the component AppleMobileFileIntegrity. The manipulation leads to improper access controls. The identification of this vulnerability is CVE-2023-27931. Attacking locally is a requirement. There is no exploit available. It is recommended to upgrade the affected component.

A vulnerability was found in Apple iOS and iPadOS up to 16.3.1 and classified as problematic. This issue affects some unknown processing of the component TCC. The manipulation leads to information disclosure. The identification of this vulnerability is CVE-2023-27931. Local access is required to approach this attack. There is no exploit available. It is recommended to upgrade the affected component.

A vulnerability was found in llvm-project a0138390. It has been classified as critical. This affects an unknown part. The manipulation leads to reachable assertion. This vulnerability is uniquely identified as CVE-2023-29935. It is possible to launch the attack on the local host. There is no exploit available.

A vulnerability was found in llvm-project 6c01b5c. It has been rated as problematic. This issue affects the function mlir::Type::getDialect. The manipulation leads to memory corruption. The identification of this vulnerability is CVE-2023-29934. An attack has to be approached locally. There is no exploit available.

A vulnerability, which was classified as critical, has been found in llvm-project a0138390. Affected by this issue is the function mlir::spirv::TargetEnv::TargetEnv(mlir::spirv::TargetEnvAttr). The manipulation leads to memory corruption. This vulnerability is handled as CVE-2023-29939. It is possible to launch the attack on the local host. There is no exploit available.

A vulnerability, which was classified as critical, was found in llvm-project a0138390. This affects the function mlir::Type::isa<mlir::LLVM::LLVMVoidType. The manipulation leads to memory corruption. This vulnerability is uniquely identified as CVE-2023-29942. The attack needs to be approached locally. There is no exploit available.