Tag: indirect prompt injection
12 articles
Web Content Conceals Hidden Instructions Targeting AI Agents
As AI agents increasingly interact with the web, hidden instructions embedded in online content can be manipulated to perform unintended actions, posing a new threat to users. Researchers have uncovered real-world campaigns that use indirect prompt injection to steer AI agents into carrying out malicious tasks.
AI Browsers Exposed to Credential-Leaking BioShocking Attack
A shocking new attack has been discovered that can trick AI browsers and assistants into leaking sensitive user credentials, with six popular agents already proven vulnerable. This sneaky tactic, called BioShocking, uses a clever game-like approach to bypass safety protocols and get agents to cough up personal info.
Researchers Expose AI Agents to Malicious Prompt Injection Payloads
Imagine a browser AI that can summarize web pages, but with a hidden vulnerability that allows malicious instructions to be embedded and executed - a newly discovered threat that security researchers are warning deserves our attention. Forcepoint researchers have uncovered 10 real-world examples of indirect prompt injection payloads designed to subvert AI agents and wreak havoc.
Researchers bypass Grafana AI with stealthy data exfiltration technique
Imagine a tool meant to reveal operational insights being turned into a stealthy spy, siphoning off sensitive corporate secrets - that's what happened when researchers exploited Grafana's AI with a cunning technique called indirect prompt injection. Dubbed GrafanaGhost, this attack bypasses Grafana's defenses, exfiltrating data without leaving a digital trail.
The Promptware Kill Chain: Exclusive Critical Risk Guide
What if a stray calendar event or shared doc could become a command to your AI? This guide reveals the promptware kill chain—how attackers weaponize language to steal data, gain persistence, and trigger unauthorized actions, and what you can do to defend against it.
Threat Actors Use Stunning, Dangerous Calendar Subs
Think that calendar invite is safe? Threat actors are weaponizing calendar subscriptions—slipping phishing links, malware, or hidden instructions into benign-seeming invites hosted on trusted services, turning everyday convenience into a stealthy breach vector.
Gemini AI Exclusive: Dangerous Thinking Robot Malware
What if the AI meant to amplify our thinking could be turned into thinking robot malware that rewrites itself to hide from defenders? New research shows attackers chaining prompt- and log-injection tricks to weaponize Gemini into self-modifying, persistent surveillance agents that sidestep many standard safeguards.
Sneaky Mermaid attack: Exclusive Copilot data breach alert
A clever Sneaky Mermaid indirect prompt injection showed how hidden instructions buried in files could trick Microsoft 365 Copilot into leaking tenant data. Microsoft says it patched this specific flaw, but security teams warn the broader risk of stealthy, embedded prompt attacks is far from over.
Microsoft 365 Copilot Exclusive: Dangerous Mermaid Attack
The Mermaid attack revealed how a hidden prompt in an otherwise harmless file could trick Microsoft 365 Copilot into spilling emails and attachments. Microsoft patched the gap, but the episode is a clear reminder that giving AI broad access can turn convenience into a new, exploitable data risk.
Sneaky Mermaid attack: Exclusive critical Copilot leak
Researchers uncovered a Sneaky Mermaid trick that hid malicious instructions inside ordinary files to make Microsoft 365 Copilot leak tenant emails and attachments. Microsoft patched the specific vector, but the episode is a wake-up call about how AI assistants can be manipulated and why teams must shore up their digital defenses.
indirect prompt injection: Stunning Risk Exposed
A trio of vulnerabilities in Google’s Gemini shows how indirect prompt injection—hiding instructions in files, metadata or chained APIs—can trick AI into leaking data or taking unintended actions, proving that securing models means vetting every input source, not just user prompts.
indirect prompt injection: Stunning, Risky Threat
Imagine a calendar invite or shared doc quietly telling your phone assistant to betray you — researchers show indirect prompt injection turns everyday interactions into real attack paths that can leak data, send messages, or trigger devices. Their TARA framework and practical fixes show those risks can fall sharply if developers add source checks, action gating, and clearer user consent.