CISA and NIST have released new technical guidance to harden cloud identity tokens against theft, forgery, and misuse in SSO and API environments.
Cyber espionage and criminal groups are increasingly targeting proprietary AI models, source code, and API credentials to fuel automated attacks and unauthorised AI workloads.
A new hardware attack called DDRop allows attackers to bypass memory encryption in Intel and AMD confidential computing environments by silently dropping memory writes.
A mass-scanning campaign is targeting internet-exposed Vite development servers to exfiltrate cloud credentials and environment configuration files.
A human attacker exploited a remote code execution vulnerability in Marimo notebooks to reach an SSH bastion host in just eight seconds using custom scripts.