Aggregator
This AI model backdoor attack stays hidden until you customize the model
Most teams that deploy AI start with a backbone model. They download a large pre-trained system, adapt it to a specific task, and put it into production. The download step carries a security question: the origin of the model. A research team built an attack called BadBone. It plants a backdoor inside a backbone model. Downstream tasks that adapt the model inherit the backdoor. The name points at the target. Corrupt the skeleton, and systems … More →
The post This AI model backdoor attack stays hidden until you customize the model appeared first on Help Net Security.
JVN: Windowsカーネルドライバ「PCTCore64.sys」における不適切なアクセス制御
美国中央情报局中国任务中心的机构设置图
Цифровая изоляция и дефицит кадров. Российский кибербез вырос до 364 миллиардов рублей, но работать там некому
黑客利用 Meta AI 机器人接管 Instagram 名人账号
Critical Security Flaw Exposes Apache LDAP API Connections
A dangerous new flaw has disrupted open-source identity infrastructure this week. Specifically, a severe Apache LDAP API vulnerability leaves directory clients vulnerable to interception. This specific framework provides an enhanced alternative to traditional Java...
The post Critical Security Flaw Exposes Apache LDAP API Connections appeared first on Information Security News.
Cybersecurity jobs available right now: June 2, 2026
Agentic Safety and Ecosystem Architect, Trust and Safety Google | USA | On-site – View job details As an Agentic Safety and Ecosystem Architect, Trust and Safety, you will define safety controls and permission models for autonomous agents on Android, helping ensure actions are reviewed before execution and access to sensitive data requires explicit consent. You will develop monitoring mechanisms to detect unsafe agent behavior and work with the developer community to promote secure agent … More →
The post Cybersecurity jobs available right now: June 2, 2026 appeared first on Help Net Security.