Advertisement
Encrypted Prompts Bypass AI Safety Guardrails in Grok and Gemini
Researchers discover cryptographic context injection, a novel technique bypassing AI safety filters in xAI Grok and Google Gemini using encryption.
Google Gemini Hijack: Command Injection via Messaging Notifications
Researchers demonstrate how Google Gemini voice assistant can be hijacked via malicious messaging notifications to control smart homes and start video calls.
Google Gemini Hijacked on Android via Poisoned Notifications
Researchers demonstrate how WhatsApp and Slack notifications can trigger indirect prompt injection in Google Gemini, leading to memory poisoning.
Google Gemini Indirect Prompt Injection via Malicious Notifications
Security researchers demonstrate how malicious notifications can manipulate Google Gemini's voice assistant to perform unauthorized tasks or exfiltrate data.
Google Gemini CLI Host Code Execution: Securing AI Developer Tools
Critical security flaw in Google Gemini CLI allows host code execution and supply chain attacks via malicious configurations. Learn how to mitigate.
Gemini CLI Critical RCE Fix: Patching the @google/gemini-cli Flaw
Google patches a CVSS 10.0 flaw in Gemini CLI tools that allowed unprivileged attackers to execute commands in CI/CD environments via malicious configurations.
Advertisement
Google Deploys Gemini AI to Combat Malvertising and Brand Fraud
Google expands the use of Gemini LLMs to detect sophisticated ad scams, blocking 5.5 billion ads and countering AI-generated brand impersonation tactics.
Google Gemini Side Panel Bug Enables Session Hijacking — Update Now
Researchers discovered a security flaw in the Google Gemini side panel that allows for unauthorized session hijacking and cross-origin data exfiltration.
Data Poisoning Risks in Real-Time AI Search and Ingestion
A recent experiment highlights how rapid web scraping for AI models like Gemini and ChatGPT enables data poisoning attacks through unverified web content.