A prompt injection flaw in Google Gemini’s voice assistant let attackers hide malicious commands in notifications, enabling social engineering and more.
Tag: prompt
Global Security News
Malicious Notifications Could Trick Google Gemini Users
A prompt injection flaw in Google Gemini’s voice assistant let attackers hide malicious commands in notifications, enabling social engineering and more.
AI, Apps, Global Security News
Microsoft Scout agent opens a new category of always-on Autopilots
Workplace AI assistants have mostly waited for a prompt before doing anything. A user asks, the tool answers, and the exchange ends there. Microsoft is putting a different kind of agent inside its Office applications, one designed to keep operating in the background once a person stops paying attention. The company introduced Microsoft Scout, calling…
AI, Cybersecurity, Global Security News
AI-generated reporting: Lessons learned from Cisco Talos Incident Response
Talos IR shares prompt engineering lessons for consistent, accurate AI-generated cybersecurity reports with a realistic case study.
AI, Apps, Exploits, Global Security News, Risk Management
New image-based prompt injection attack targets multimodal AI models
Security researchers have developed a new image-based prompt injection attack that can manipulate how multimodal AI systems interpret user instructions without modifying the original text prompt, potentially expanding security risks for AI agents and vision-language systems. In a research paper published this week, researchers from Xidian University described a technique called “CrossMPI,” which uses nearly…
AI, Global Security News
Google Fixes CVSS 10 Gemini CLI Vulnerability Enabling GitHub Issue-Based RCE
Google patches a CVSS 10 Gemini CLI vulnerability that allowed hackers to use prompt injection and privilege escalation for a full supply chain compromise.
AI, Cybersecurity, Global Security News, Government & Policy, Risk Management
Security agencies draw red lines around agentic AI deployments
With prompt injection and other attack pathways consistently surfacing across agentic AI deployments, security watchdogs have stepped in, collectively, to draw some hard boundaries. A joint advisory from the US Cybersecurity and Infrastructure Security Agency (CISA) and international partners has called for tighter control over permissions, stronger monitoring, and a more deliberate rollout strategy, urging…
AI, Apps, Cybersecurity, Global Security News, Network Security, Risk Management
Stopping the quiet drift toward excessive agency with re-permissioning
In their infancy, LLM models were not difficult to contain. You gave a prompt; they responded, and if something was wrong it was usually “just text.” This could take the form of a summary that missed the best bits, a tone-deaf line or a wordy sentence. But then, agents were co-opted as the core reasoning…
AI, Cybersecurity, Exploits, Global Security News
Hackers Use Hidden Website Instructions in New Attacks on AI Assistants
Cybersecurity researchers at Forcepoint uncover new indirect prompt injection attacks that use hidden website code to exploit AI assistants like GitHub Copilot.
AI, Global Security News
Researchers Uncover 10 In-the-Wild Prompt Injection Payloads Targeting AI Agents
Forcepoint has found 10 new indirect prompt injection attacks targeting AI agents
AI, Global Security News
Google Fixes Critical RCE Flaw in AI-Based Antigravity Tool
The prompt injection vulnerability in the agentic AI product for filesystem operations was a sanitization issue that allowed for sandbox escape and arbitrary code execution.
AI, Exploits, Global Security News, Network Security
Prompt injection turned Google’s Antigravity file search into RCE
Security researchers have revealed a prompt injection flaw in Google’s Antigravity IDE that could be weaponized to bypass its sandbox protections and achieve remote code execution (RCE). The issue came from Antigravity’s ability to allow AI agents to invoke native functions, like searching files, on behalf of the user. Designed to kill complexity, the feature…
Global Security News
Microsoft, Salesforce Patch AI Agent Data Leak Flaws
Two recently fixed prompt injections in Salesforce Agentforce and Microsoft Copilot would have enabled an external attacker to leak sensitive data.
AI, Global Security News
GrafanaGhost Vulnerability Allows Data Theft via AI Injection
GrafanaGhost is a critical vulnerability in Grafana’s AI components that uses indirect prompt injection and protocol-relative URL bypasses to exfiltrate data.
AI, Global Security News
GrafanaGhost Exploit Bypasses AI Guardrails for Silent Data Exfiltration
GrafanaGhost chains AI prompt injection and URL flaws to exfiltrate sensitive Grafana data
AI, Apps, Exploits, Global Security News, Network Security, Risk Management
Zero‑click Grafana AI attack can enable enterprise data exfiltration
Indirect prompt injection is possible on AI-powered dashboards, allowing exfiltration of sensitive enterprise data without user authentication. Security researchers are warning about a critical Grafana issue, dubbed GrafanaGhost, that allows attackers to leak sensitive data from Grafana environments, including financial metrics, infrastructure health data, private customer data, and operational logs, among others. Noma Security disclosed…
AI, Apps, Exploits, Global Security News
Google Workspace’s continuous approach to mitigating indirect prompt injections
Posted by Adam Gavish, Google GenAI Security Team Indirect prompt injection (IPI) is an evolving threat vector targeting users of complex AI applications with multiple data sources, such as Workspace with Gemini. This technique enables the attacker to influence the behavior of an LLM by injecting malicious instructions into the data or tools used by…
AI, Data Breaches, Exploits, Global Security News, Network Security, Risk Management
Check Point Research Reveals ChatGPT Data Exfiltration Flaw
A flaw in ChatGPT’s code execution environment shows how a single malicious prompt could quietly leak sensitive user data — without any warning or user approval needed. “Sensitive data shared with ChatGPT conversations could be silently exfiltrated without the user’s knowledge or approval,” said Check Point researchers. Inside the ChatGPT DNS Exfiltration Flaw The issue…
Global Security News
A nearly undetectable LLM attack needs only a handful of poisoned samples
Prompt engineering has become a standard part of how large language models are deployed in production, and it introduces an attack surface most organizations have not yet addressed. Researchers have developed and tested a prompt-based backdoor attack method, called ProAttack, that achieves attack success rates approaching 100% on multiple text classification benchmarks without altering sample…
AI, Apps, Exploits, Global Security News, Risk Management
Microsoft details AI prompt abuse techniques targeting AI assistants
Prompt abuse occurs when crafted inputs manipulate an AI system into producing unintended behavior, such as attempting to access sensitive information or overriding built-in safety instructions. Prompt injection is also recognized as one of the top risks in the 2025 OWASP guidance for LLM applications. “Detecting abuse is challenging because it exploits natural language, such…
AI, Apps, Global Security News
Arcjet enables inline defense against prompt injection in production AI systems
Arcjet has released AI Prompt Injection Protection, a new capability designed to stop prompt injection attacks before they reach production AI models. The feature detects hostile prompts at the application boundary and gives developers a decision point inside the request lifecycle where malicious instructions can be blocked before inference occurs. Companies are shipping AI features…
AI, Global Security News, Network Security
‘Claudy Day’ Trio of Flaws Exposes Claude Users to Data Theft
A prompt injection vulnerability paired with other flaws can turn a Google search into a full attack chain that could threaten enterprise networks.
AI, Apps, Global Security News, Risk Management
Mend.io eliminates AI prompt weaknesses before production
Mend.io has launched System Prompt Hardening within Mend AI to detect, score, and automatically remediate weaknesses in AI system prompts. Hidden instructions in system prompts have emerged as a growing security concern that traditional AppSec tools do not fully address. System Prompt Hardening provides instant visibility into these behind-the-scenes instructions, identifies weaknesses, and automatically strengthens…
AI, Global Security News, Government & Policy
Cyberattack on Mexico’s Gov’t Agencies Highlight AI Threat
Using Anthropic’s Claude, OpenAI’s ChatGPT, and a detailed playbook prompt, a handful of cyberattackers reportedly gained access to government agencies and its citizens’ data.
AI, Artificial Intelligence, Generative AI, Security, Cybersecurity, Data Breaches, Exploits, Global Security News, Risk Management
Single prompt breaks AI safety in 15 major language models
A single benign-sounding prompt can systematically strip safety guardrails from major language and image models, raising fresh questions about the durability of AI alignment when models are customized for enterprise use, according to Microsoft research. The technique, dubbed GRP-Obliteration, weaponizes a common AI training method called Group Relative Policy Optimization, normally used to make models…