Latest cybersecurity reporting from selected sources.

Anthropic fixed the flaws - but the AI-enabled attack surfaces remain Security vulnerabilities in Claude Code could have allowed attackers to remotely execute code on users' machines and steal API keys by injecting malicious configurations into repositories, and then waiting for a developer to clone and open an untrustworthy project.…

Researcher shows how agentic AI is vulnerable to hijacking to subvert an agent's goals and how agent interaction can be altered to compromise whole networks.

Attackers are using AI to weaponize old vulnerabilities while security teams face expanding attack surfaces and limited resources. Intruder's 2025 Exposure Management Index reveals how 3,000+ organizations are adapting and fixing critical flaws faster than ever. [...]

Lowering Machine Identity Risks in AI, ML and Bot WorkflowsWhile AI, ML and bot workflows boost efficiency, they also expand the attack surface. Over-permissioned identities, exploitable vulnerabilities and AI misuse pose significant security risks. AI-driven security tools can mitigate these risks by detecting anomalies and automating threat response.

RSA's Innovation Sandbox 2023 focused on the software supply chain, as well as attack surfaces exposed by generative AI, ML systems, and APIs.