Tag: Prompt Injection

Metabase Releases Security Updates After Zero Day Vulnerability Grants Unauthorized Administrator Access

Risk & Resilience
Metabase has released security updates after confirming active exploitation of a zero day vulnerability that allows attackers to gain administrator access without authentication.

Black Hat 2026 Research Reveals CSS Techniques That Can Target Webmail Passwords And Authentication Tokens

Risk & Resilience
Research presented at Black Hat USA 2026 demonstrates CSS based attack techniques affecting major webmail services, highlighting risks involving passwords, authentication tokens, and AI connected email workflows.

Claude Code GitHub Action Flaw Exposed Public Repositories To Repository Hijacking Risks

Risk & Resilience
A security flaw in Anthropic’s Claude Code GitHub Action allowed attackers to potentially hijack public repositories through a single malicious GitHub issue.

WhatsApp And Slack Notifications Could Manipulate Google Gemini On Android Devices

Risk & Resilience
Security researchers have disclosed how poisoned notifications from apps like WhatsApp and Slack could manipulate Google Gemini on Android, potentially triggering unauthorized actions and memory changes.

Google Patches Antigravity IDE Flaw As Researchers Expose Expanding Prompt Injection Attack Surface In AI Tools

Risk & Resilience
Google fixes Antigravity IDE vulnerability enabling prompt injection based code execution as researchers uncover wider AI tool security flaws across coding agents and platforms.

Secure by Design: Why AI Governance Can No Longer Trail AI Deployment

Risk & Resilience
AI governance must evolve alongside deployment as agentic systems expand enterprise attack surfaces. This analysis explores secure-by-design frameworks, emerging AI threats, and why Pakistani enterprises must strengthen internal governance amid rapid AI adoption and fragmented regulation.

OpenClaw AI Agent Security Flaws Raise Prompt Injection And Data Exfiltration Concerns

Risk & Resilience
China’s CNCERT warns that OpenClaw AI agent security weaknesses could enable prompt injection attacks, endpoint compromise, and sensitive data exfiltration.

When Machines Read the Internet: What Security Champions Need to Know About Prompt Injection

Editorial Insights
As enterprises deploy AI agents that read and act on information from internal systems and the internet, prompt injection is emerging as a new cybersecurity risk that can manipulate machine reasoning, expose sensitive data, and influence automated workflows.

Microsoft Warns Of Manipulative Prompts Hidden In Summarize With AI Buttons

Risk & Resilience
Microsoft researchers identify more than 50 hidden prompts embedded in Summarize with AI buttons that influence assistants to remember and recommend specific brands without user awareness.

Recent articles

spot_img