Agentic AI expands your attack surface because agents read and act on untrusted content. Consequently, indirect prompt injection can hijack tool use, leak data, and trigger risky actions. This guide explains how the attack works, how to detect it, and how to deploy guardrails that actually help at enterprise scale.
TEE.Fail uses a DDR5 interposer to undermine enclave confidentiality on Intel TDX and AMD SEV-SNP. Because memory encryption lacks strong integrity here, defenders should rethink secrets, strengthen attestation, and tighten physical controls.
The Oracle E-Business Suite campaign continues to grow. This analysis explains the expanding victim list, enterprise impact, and the steps teams should take now to patch, hunt, and contain risk.
Attackers abused a Chrome zero-day to install Memento spyware in targeted operations. This analysis explains the chain, highlights reliable signals to hunt, and outlines a focused 72-hour action plan.
A former general manager at a top-tier cyber-weapons developer faces criminal charges after allegedly stealing eight trade secrets between 2022 and 2025 and selling them to a Russian buyer. The case underscores serious risks to national security, supply-chain oversight, and insider threat policies across the defense-cyber industry.
Qilin ransomware now combines a Linux payload with a BYOVD (Bring-Your-Own-Vulnerable-Driver) exploit, enabling affiliates to bypass endpoint controls and compromise virtualised and Windows environments. This briefing explains the attack chain, detection challenges, and immediate defensive steps security teams must apply.
Security researchers revealed that ChatGPT’s Atlas Browser can be manipulated through hidden prompt injections, allowing attackers to hijack AI behavior, leak data, and bypass safeguards. Learn how it works and how to defend against it.
The latest version of the ransomware family known as LockBit has resurfaced with a potent new variant, LockBit 5.0, capable of striking Windows endpoints, Linux servers and VMware ESXi hypervisor platforms in one campaign. Organisations must reassess their ransomware defences and detection posture now.
A newly weaponized version of the open-source toolkit RedTiger is actively targeting Discord accounts and gaming credentials. This article examines the malware’s mechanics, why gamers and enterprises should care, and outlines a mitigation roadmap.
The Lazarus Group launched a sophisticated social-engineering campaign targeting European unmanned aerial vehicle (UAV) manufacturers via fake job offers to steal intellectual property and design data. Understanding the “Operation DreamJob” method, infection chain and mitigation is critical for defense supply-chain security.