A security analyst reviewing a ticket about an AI tool's wrong recommendation reveals a critical gap: risk registers document AI risks but cannot answer who has authority to stop a system, according t…
Amazon Web Services chief security officer Steve Schmidt says AI has compressed security work that once took months into minutes, with detections now built in about 15 minutes. However, security leade…
CrowdStrike identified five new prompt injection techniques that exploit AI systems by tricking large language models into accepting malicious instructions. The attacks include trigger-activated rule …
Check Point Software CTO Jonathan Zanger said AI is driving the biggest change in cybersecurity since the internet, enabling defenders to scale threat detection while also empowering attackers with fa…
Cybersecurity firm Wiz discovered GhostApproval, a vulnerability affecting six major AI coding assistants including Amazon Q Developer, Anthropic Claude Code, and Cursor, that allows attackers to esca…
Attackers compromised an AWS EC2 instance acting as a LiteLLM proxy for Amazon Bedrock, deploying XMRig cryptomining malware and attempting to abuse cloud identities and AI services. Cybersecurity fir…
The UK's National Cyber Security Centre (NCSC) unveiled Cyber Shield, a blueprint for deploying autonomous AI agents to detect and neutralize cyberattacks in real time, aiming to build a machine-speed…
A new six-stage maturity model for non-human identities (NHIs) addresses the identity governance gap exposed by agentic AI, where LLM-based agents with excessive privileges have caused production outa…
Security leaders are spending billions on AI-driven cybersecurity tools, but most performance issues stem from fragmented data pipelines and schema drift rather than flawed detection models. Gartner p…
Zscaler found that several high-end AI agents, including Llama3 and Gemini models, fell victim to indirect prompt injection traps that would fool few humans, while some lower-level LLMs performed bett…
Zscaler found that several major AI models, including Llama3 and Gemini variants, are vulnerable to indirect prompt injection traps, while others like Llama4-maverick and Gemini-3.1-pro proved safe. E…
AI agents, described as 'really smart kindergartners' by HashiCorp CTO Stephen Wilson, are creating blind spots in zero trust security programs by moving too fast for traditional authentication and ac…
IBM argues that existing security controls are insufficient for AI agents, requiring a new operational control plane for agentic identity, dynamic secrets, and least-privilege access across multi-envi…
HashiCorp field CTO Stephen Wilson warns that as organizations move from using AI as assistants to autonomous agents, security and governance practices are not keeping up. Wilson outlines three adopti…
A fully autonomous AI agent named JadePuffer conducted an end-to-end cyber intrusion and extortion campaign by exploiting a vulnerable Langflow server, marking the first documented case of agentic ran…
The cybersecurity industry's perceived skills gap is actually a validation gap, as traditional training and certifications fail to keep pace with AI-driven threats. Organizations lack hands-on readine…
Security firm AIR demonstrated a fake AI agent skill that bypassed all static security scanners and reached over 26,000 Instagram users, highlighting how AI skills can change behavior after approval a…
Adobe announced it will issue security patches twice monthly, starting July 14, to address the increasing pace of vulnerability discovery and exploitation driven by AI. The company will add a second P…
Researchers discovered two sandbox bypass vulnerabilities in Cursor IDE that allow remote code execution via prompt injection. Tracked as CVE-2026-50548 and CVE-2026-50549, the flaws let attackers bre…
Google removed a malicious Chromium extension spoofing Perplexity AI after Microsoft researchers found it intercepted users' search traffic and routed queries through attacker-controlled servers. The …