Cybersecurity articles tagged #Prompt Injection on RuntimeRebel.https://runtimerebel.comhttps://runtimerebel.com/blog/autonomous-ai-agents-email-security-insights-on-perimeter-defenceshttps://runtimerebel.com/blog/autonomous-ai-agents-email-security-insights-on-perimeter-defencesAn autonomous AI agent emailed security researcher Bruce Schneier detailing perimeter defences, anti-bot mechanisms, and invisible prompt injections.Wed, 02 Sep 2026 19:09:10 GMTArtificial IntelligencePrompt InjectionBot ManagementEmail Securityhttps://runtimerebel.com/blog/the-agentic-soc-from-ai-theater-to-real-defensehttps://runtimerebel.com/blog/the-agentic-soc-from-ai-theater-to-real-defenseExplore the Agentic SOC transition, focusing on measurable AI ROI, new risks like indirect prompt injection, and redefining analyst roles for autonomous defense.Tue, 01 Sep 2026 19:03:18 GMTAI in SecuritySOC OperationsPrompt InjectionAutomationThreat Intelligencehttps://runtimerebel.com/blog/ai-email-summarizers-vulnerable-to-hidden-html-promptshttps://runtimerebel.com/blog/ai-email-summarizers-vulnerable-to-hidden-html-promptsAttackers can use invisible HTML prompts to manipulate AI email summarizers, generating false information and potential security risks.Wed, 26 Aug 2026 00:44:15 GMTAIPrompt InjectionEmail SecuritySocial EngineeringHidden HTMLhttps://runtimerebel.com/blog/alice-secures-140m-to-enhance-ai-model-defenses-and-guardrailshttps://runtimerebel.com/blog/alice-secures-140m-to-enhance-ai-model-defenses-and-guardrailsAI security firm Alice raised $140M to combat adversarial AI, prompt injection, and jailbreak attempts in generative AI systems.Wed, 26 Aug 2026 00:43:29 GMTAI SecurityAdversarial AIPrompt InjectionGenerative AIFundinghttps://runtimerebel.com/blog/encrypted-prompts-bypass-ai-safety-guardrails-in-grok-and-geminihttps://runtimerebel.com/blog/encrypted-prompts-bypass-ai-safety-guardrails-in-grok-and-geminiResearchers discover cryptographic context injection, a novel technique bypassing AI safety filters in xAI Grok and Google Gemini using encryption.Sat, 22 Aug 2026 08:19:25 GMTArtificial IntelligencePrompt InjectionZero-DayData BreachGoogle Geminihttps://runtimerebel.com/blog/cryptographic-context-injection-exposes-grok-chat-datahttps://runtimerebel.com/blog/cryptographic-context-injection-exposes-grok-chat-dataAdversa AI reveals Cryptographic Context Injection, allowing web pages to steal Grok user data and chat prompts without user consent.Thu, 20 Aug 2026 16:23:24 GMTXaiAdversa AIPrompt InjectionData ExfiltrationGrokhttps://runtimerebel.com/blog/cve-2026-24301-cosnitch-exploits-microsoft-copilot-personalhttps://runtimerebel.com/blog/cve-2026-24301-cosnitch-exploits-microsoft-copilot-personalVaronis disclosed CoSnitch (CVE-2026-24301), affecting Microsoft Copilot Personal, enabling one-click data exfiltration and persistent memory poisoning.Wed, 19 Aug 2026 00:39:19 GMTData ExfiltrationPrompt InjectionCVE-2026-24301Microsoft Copilot PersonalMemory Poisoninghttps://runtimerebel.com/blog/ai-mind-viruses-spread-via-persistent-prompt-fileshttps://runtimerebel.com/blog/ai-mind-viruses-spread-via-persistent-prompt-filesSecurity research reveals self-propagating AI 'mind viruses' can spread between autonomous agents through editable system prompt files.Tue, 18 Aug 2026 16:19:59 GMTAILarge Language ModelsPrompt InjectionAnthropicAutonomous Agentshttps://runtimerebel.com/blog/context-bombing-defending-against-ai-hacking-agentshttps://runtimerebel.com/blog/context-bombing-defending-against-ai-hacking-agentsResearchers at Tracebit introduce 'context bombing,' a defensive prompt injection technique to shut down AI hacking agents by exploiting guardrails.Wed, 12 Aug 2026 16:50:19 GMTAI SecurityPrompt InjectionLarge Language ModelsAWS SecurityTracebithttps://runtimerebel.com/blog/anthropic-opus-5-significantly-boosts-prompt-injection-resistancehttps://runtimerebel.com/blog/anthropic-opus-5-significantly-boosts-prompt-injection-resistanceAnthropic's Opus 5 demonstrates superior resistance to prompt injection attacks on the IPI benchmark, outperforming other leading LLMs, including GPT-5.6 variants.Sat, 01 Aug 2026 10:01:46 GMTAnthropic Opus 5Prompt InjectionLLM SecurityIPI BenchmarkAI Securityhttps://runtimerebel.com/blog/claude-mythos-securing-llms-in-enterprise-hype-vs-realityhttps://runtimerebel.com/blog/claude-mythos-securing-llms-in-enterprise-hype-vs-realityExamine the security implications of Anthropic's Claude Mythos and other LLMs in enterprise.Thu, 30 Jul 2026 17:32:21 GMTClaude MythosAnthropicLarge Language ModelsLLM SecurityPrompt InjectionAI GovernanceEnterprise AIhttps://runtimerebel.com/blog/persistent-prompt-injection-risks-in-microsoft-365-copilot-for-wordhttps://runtimerebel.com/blog/persistent-prompt-injection-risks-in-microsoft-365-copilot-for-wordResearchers demonstrate how hidden instructions in Word documents can persist through Microsoft 365 Copilot drafting, creating risks of malicious prompt propagation.Thu, 30 Jul 2026 14:05:35 GMTMicrosoft 365CopilotPrompt InjectionAI SecurityDocument Securityhttps://runtimerebel.com/blog/plc-exploits-and-ai-prompt-injection-analysis-of-emerging-threatshttps://runtimerebel.com/blog/plc-exploits-and-ai-prompt-injection-analysis-of-emerging-threatsAnalysis of recent threats including PLC attacks, AI image prompt injection, and Android spyware disguised as utility applications in the latest bulletin.Thu, 23 Jul 2026 17:26:37 GMTPLC SecurityAI SecurityAndroid SpywarePrompt InjectionICS Securityhttps://runtimerebel.com/blog/aws-kiro-rce-via-indirect-prompt-injection-mitigation-guidehttps://runtimerebel.com/blog/aws-kiro-rce-via-indirect-prompt-injection-mitigation-guideResearch reveals a critical flaw in AWS Kiro where malicious web pages trigger RCE by rewriting configuration files via indirect prompt injection attacks.Tue, 21 Jul 2026 17:22:23 GMTAWSKiroRCEPrompt InjectionAI Securityhttps://runtimerebel.com/blog/open-source-android-ai-agent-hijacking-leads-to-host-system-rcehttps://runtimerebel.com/blog/open-source-android-ai-agent-hijacking-leads-to-host-system-rceLearn how invisible text exploits open-source Android AI agents to trigger malicious code execution on host PCs via indirect prompt injection.Tue, 21 Jul 2026 13:54:29 GMTAndroid SecurityAI AgentsPrompt InjectionRCEMobile SecurityAppagenthttps://runtimerebel.com/blog/openai-gpt-red-automating-prompt-injection-discovery-for-gpt-5-6-solhttps://runtimerebel.com/blog/openai-gpt-red-automating-prompt-injection-discovery-for-gpt-5-6-solOpenAI reveals GPT-Red, an automated red-teaming model designed to detect prompt injection vulnerabilities and harden GPT-5.6 Sol via adversarial training.Thu, 16 Jul 2026 10:10:18 GMTOpenAIGPT RedGPT 5 6 SolPrompt InjectionAdversarial Traininghttps://runtimerebel.com/blog/promptfiction-claude-ai-vulnerability-exploits-malicious-promptshttps://runtimerebel.com/blog/promptfiction-claude-ai-vulnerability-exploits-malicious-promptsDiscover PromptFiction, a fixed vulnerability in Claude AI that allowed malicious prompts to trigger end-to-end attacks. Learn mitigation for AI agent security.Wed, 15 Jul 2026 17:22:45 GMTPromptFictionClaude AIAI SecurityPrompt InjectionVulnerabilityhttps://runtimerebel.com/blog/memghost-attack-persistent-memory-poisoning-in-ai-agents-via-emailhttps://runtimerebel.com/blog/memghost-attack-persistent-memory-poisoning-in-ai-agents-via-emailThe MemGhost attack targets AI agent memory systems via email, planting persistent false facts to stealthily manipulate long-term assistant behavior.Mon, 13 Jul 2026 17:59:10 GMTMemGhostAI SecurityPrompt InjectionLLM AgentsAdversarial MLhttps://runtimerebel.com/blog/ghostcommit-hidden-prompt-injection-in-images-targets-ai-agentshttps://runtimerebel.com/blog/ghostcommit-hidden-prompt-injection-in-images-targets-ai-agentsResearchers demonstrate Ghostcommit, a technique using images to hide prompt injection attacks that trick AI agents into exfiltrating repository secrets.Sat, 11 Jul 2026 09:39:43 GMTGhostcommitPrompt InjectionSteganographyAI SecurityCoderabbitBugbothttps://runtimerebel.com/blog/how-agentjacking-exploits-ai-coding-agents-via-fake-bug-reportshttps://runtimerebel.com/blog/how-agentjacking-exploits-ai-coding-agents-via-fake-bug-reportsResearchers demonstrate 'Agentjacking,' a technique using indirect prompt injection to hijack AI coding agents through malicious GitHub bug reports.Wed, 01 Jul 2026 09:21:10 GMTAgentjackingPrompt InjectionAI SecurityTrail of BitsOpendevinSupply Chain Securityhttps://runtimerebel.com/blog/ai-agents-vulnerable-to-data-leak-via-poisoned-mcp-toolshttps://runtimerebel.com/blog/ai-agents-vulnerable-to-data-leak-via-poisoned-mcp-toolsMicrosoft warns that malicious tool descriptions for AI agents can lead to stealthy data exfiltration, bypassing security controls by mimicking routine actions.Wed, 01 Jul 2026 01:02:51 GMTAI AgentsData ExfiltrationSupply Chain AttackMicrosoftPrompt Injectionhttps://runtimerebel.com/blog/claude-code-indirect-prompt-injection-hijacking-developer-machineshttps://runtimerebel.com/blog/claude-code-indirect-prompt-injection-hijacking-developer-machinesResearchers demonstrate a new attack method leveraging indirect prompt injection in Claude Code, enabling the hijack of developer machines via malicious code in…Mon, 29 Jun 2026 17:07:49 GMTClaude CodePrompt InjectionSupply Chain AttackDeveloper SecurityAI SecurityReverse ShellMachine Hijackhttps://runtimerebel.com/blog/llm-prompt-injection-role-confusion-exposes-core-architectural-flawshttps://runtimerebel.com/blog/llm-prompt-injection-role-confusion-exposes-core-architectural-flawsAn in-depth analysis of LLM prompt injection, detailing how 'role confusion' in model representations undermines tag-based security and demands architectural solutions.Thu, 25 Jun 2026 13:06:12 GMTPrompt InjectionLLM SecurityAI SecurityRole ConfusionLarge Language Modelshttps://runtimerebel.com/blog/gaslight-macos-malware-uses-prompt-injection-to-bypass-ai-detectionhttps://runtimerebel.com/blog/gaslight-macos-malware-uses-prompt-injection-to-bypass-ai-detectionGaslight is a newly discovered Rust-based macOS implant that uses prompt injection to deceive AI-driven analysis tools and bypass automated detection.Thu, 25 Jun 2026 13:01:04 GMTGaslight MalwaremacOS SecurityPrompt InjectionAnti AnalysisRust Malwarehttps://runtimerebel.com/blog/ai-agent-traps-information-as-an-attack-surface-for-autonomous-systemshttps://runtimerebel.com/blog/ai-agent-traps-information-as-an-attack-surface-for-autonomous-systemsAttackers exploit trusted data sources to deploy AI agent traps, leading to hidden content injections and cognitive state poisoning for autonomous AI systems.Thu, 25 Jun 2026 00:59:45 GMTAI SecurityAI Agent TrapsData PoisoningLarge Language ModelsAutonomous AIAttack SurfacePrompt Injectionhttps://runtimerebel.com/blog/bypass-ai-malware-scanners-via-policy-triggering-prompt-injectionhttps://runtimerebel.com/blog/bypass-ai-malware-scanners-via-policy-triggering-prompt-injectionMalware authors are embedding 'forbidden' text into code to trigger safety refusals in AI-mediated security scanners, effectively bypassing automated analysis.Fri, 19 Jun 2026 09:50:19 GMTAI BypassAdversarial AIPrompt InjectionMalware ObfuscationBioinformaticshttps://runtimerebel.com/blog/microsoft-copilot-searchleak-attack-ai-prompt-injection-data-thefthttps://runtimerebel.com/blog/microsoft-copilot-searchleak-attack-ai-prompt-injection-data-theftAnalysis of the critical Microsoft Copilot 'SearchLeak' attack. Learn how prompt injection allowed 1-click data theft and crucial defense strategies for AI applications.Tue, 16 Jun 2026 13:59:42 GMTMicrosoft CopilotAI SecurityPrompt InjectionSearchLeakData TheftLLM Securityhttps://runtimerebel.com/blog/openai-chatgpt-lockdown-mode-mitigating-prompt-injection-exfiltrationhttps://runtimerebel.com/blog/openai-chatgpt-lockdown-mode-mitigating-prompt-injection-exfiltrationOpenAI introduces ChatGPT Lockdown Mode for personal accounts to prevent prompt injection attacks from exfiltrating sensitive data via external tools.Sat, 06 Jun 2026 16:29:48 GMTOpenAIChatGPTPrompt InjectionData ExfiltrationAI SecurityLockdown Modehttps://runtimerebel.com/blog/google-gemini-hijack-command-injection-via-messaging-notificationshttps://runtimerebel.com/blog/google-gemini-hijack-command-injection-via-messaging-notificationsResearchers demonstrate how Google Gemini voice assistant can be hijacked via malicious messaging notifications to control smart homes and start video calls.Thu, 04 Jun 2026 13:16:24 GMTGoogle GeminiPrompt InjectionVoice AssistantIot SecurityAndroid Securityhttps://runtimerebel.com/blog/google-gemini-hijacked-on-android-via-poisoned-notificationshttps://runtimerebel.com/blog/google-gemini-hijacked-on-android-via-poisoned-notificationsResearchers demonstrate how WhatsApp and Slack notifications can trigger indirect prompt injection in Google Gemini, leading to memory poisoning.Wed, 03 Jun 2026 21:09:55 GMTGoogle GeminiAndroid SecurityPrompt InjectionAI Securityhttps://runtimerebel.com/blog/google-gemini-indirect-prompt-injection-via-malicious-notificationshttps://runtimerebel.com/blog/google-gemini-indirect-prompt-injection-via-malicious-notificationsSecurity researchers demonstrate how malicious notifications can manipulate Google Gemini's voice assistant to perform unauthorized tasks or exfiltrate data.Wed, 03 Jun 2026 13:49:53 GMTGoogle GeminiPrompt InjectionLLM SecurityHiddenlayerMobile Securityhttps://runtimerebel.com/blog/evaluating-ai-agent-security-100-agents-tested-for-vulnerabilitieshttps://runtimerebel.com/blog/evaluating-ai-agent-security-100-agents-tested-for-vulnerabilitiesAn industry-first evaluation of 100 AI agents highlights critical security gaps in defense and the high impact of potential agentic compromises.Wed, 03 Jun 2026 13:49:24 GMTAI SecurityLLM VulnerabilitiesPrompt InjectionAgentic AIhttps://runtimerebel.com/blog/meta-ai-support-bot-exploited-for-instagram-account-takeovershttps://runtimerebel.com/blog/meta-ai-support-bot-exploited-for-instagram-account-takeoversHackers manipulated Meta's AI support assistant to bypass authentication and seize high-profile Instagram accounts, including government entities.Mon, 01 Jun 2026 18:08:48 GMTInstagramMeta AIAccount TakeoverPrompt InjectionSocial Engineeringhttps://runtimerebel.com/blog/chatgpt-chatgphish-vulnerability-web-summaries-lead-to-phishinghttps://runtimerebel.com/blog/chatgpt-chatgphish-vulnerability-web-summaries-lead-to-phishingA newly disclosed ChatGPhish vulnerability allows attackers to leverage ChatGPT's Markdown trust for prompt injections and sophisticated phishing campaigns.Fri, 29 May 2026 20:53:47 GMTChatGPTChatGPhishOpenAIPhishingPrompt InjectionAI Securityhttps://runtimerebel.com/blog/anthropic-claude-mythos-class-models-security-implications-of-public-rollouthttps://runtimerebel.com/blog/anthropic-claude-mythos-class-models-security-implications-of-public-rolloutAnthropic confirms public rollout plans for Claude Mythos-class models, addressing previous delays caused by software security risks and safety concerns.Fri, 29 May 2026 01:00:17 GMTAnthropicClaude MythosAI SecurityPrompt InjectionLarge Language Modelshttps://runtimerebel.com/blog/microsoft-rampart-and-clarity-securing-ai-agents-against-exploitationhttps://runtimerebel.com/blog/microsoft-rampart-and-clarity-securing-ai-agents-against-exploitationMicrosoft open-sources RAMPART and Clarity to provide developers with frameworks for red teaming and observing autonomous AI agents against prompt injection.Wed, 20 May 2026 17:10:34 GMTAI SecurityMicrosoft RAMPARTPrompt InjectionRed TeamingAI Agentshttps://runtimerebel.com/blog/claude-code-sandbox-bypass-anthropic-patches-cli-vulnerabilityhttps://runtimerebel.com/blog/claude-code-sandbox-bypass-anthropic-patches-cli-vulnerabilityAnthropic recently addressed a sandbox bypass in Claude Code. This vulnerability could have allowed data exfiltration when combined with prompt injection.Wed, 20 May 2026 13:04:49 GMTAnthropicClaude CodeSandbox BypassAI SecurityPrompt Injectionhttps://runtimerebel.com/blog/secure-ai-agent-integration-preventing-production-data-losshttps://runtimerebel.com/blog/secure-ai-agent-integration-preventing-production-data-lossOrganizations face catastrophic data loss as AI agents misinterpret prompts. Learn how to secure autonomous agents and implement strict guardrails.Fri, 01 May 2026 16:27:57 GMTArtificial IntelligenceCloud SecurityData ProtectionAI AgentsPrompt Injectionhttps://runtimerebel.com/blog/malicious-ai-prompt-injection-attacks-google-red-team-insightshttps://runtimerebel.com/blog/malicious-ai-prompt-injection-attacks-google-red-team-insightsGoogle reports a surge in AI prompt injection attacks, highlighting low-sophistication attempts and strategies for mitigating indirect prompt injection risks.Mon, 27 Apr 2026 12:49:10 GMTAI SecurityPrompt InjectionLLM VulnerabilitiesGoogle Red Teamhttps://runtimerebel.com/blog/security-risks-of-agentic-ai-in-enterprise-ecosystemshttps://runtimerebel.com/blog/security-risks-of-agentic-ai-in-enterprise-ecosystemsAnalysis of security risks in Agentic AI adoption, focusing on prompt injection, autonomous execution, and enterprise mitigation strategies.Wed, 22 Apr 2026 08:48:37 GMTAgentic AILLM SecurityPrompt InjectionEnterprise Riskhttps://runtimerebel.com/blog/google-antigravity-rce-via-prompt-injection-mitigation-guidehttps://runtimerebel.com/blog/google-antigravity-rce-via-prompt-injection-mitigation-guideGoogle patched a critical RCE flaw in its AI-based Antigravity tool, stemming from a prompt injection vulnerability allowing sandbox escape and arbitrary code execution.Tue, 21 Apr 2026 16:32:40 GMTGoogle AntigravityRCEPrompt InjectionAI SecuritySandbox Escapehttps://runtimerebel.com/blog/cursor-ai-rce-via-indirect-prompt-injection-mitigation-guidehttps://runtimerebel.com/blog/cursor-ai-rce-via-indirect-prompt-injection-mitigation-guideSecurity researchers demonstrate how indirect prompt injection in Cursor AI could lead to full shell access on developer workstations. Patch immediately.Fri, 17 Apr 2026 08:43:51 GMTCursor AiRCEPrompt InjectionDeveloper SecurityJohann Rehbergerhttps://runtimerebel.com/blog/claude-code-and-gemini-cli-prompt-injection-via-code-commentshttps://runtimerebel.com/blog/claude-code-and-gemini-cli-prompt-injection-via-code-commentsResearch reveals how Claude Code, Gemini CLI, and GitHub Copilot agents are vulnerable to prompt injection attacks via malicious source code comments.Thu, 16 Apr 2026 08:41:38 GMTClaude CodeGemini CLIGitHub CopilotPrompt InjectionAI Securityhttps://runtimerebel.com/blog/microsoft-and-salesforce-patch-prompt-injection-flaws-in-ai-agentshttps://runtimerebel.com/blog/microsoft-and-salesforce-patch-prompt-injection-flaws-in-ai-agentsResearchers identified prompt injection vulnerabilities in Microsoft Copilot and Salesforce Agentforce that could allow attackers to exfiltrate sensitive data.Wed, 15 Apr 2026 12:32:01 GMTMicrosoft CopilotSalesforce AgentforcePrompt InjectionData ExfiltrationAI Securityhttps://runtimerebel.com/blog/google-vertex-ai-security-mitigating-ai-agent-weaponizationhttps://runtimerebel.com/blog/google-vertex-ai-security-mitigating-ai-agent-weaponizationGoogle patches security flaws in Vertex AI after Unit 42 researchers demonstrated how to weaponize AI agents for unauthorized data access and exfiltration.Wed, 01 Apr 2026 08:39:46 GMTGoogle CloudVertex AiAI Agent SecurityPrompt InjectionUnit 42https://runtimerebel.com/blog/openai-model-behavior-bug-bounty-reporting-ai-safety-riskshttps://runtimerebel.com/blog/openai-model-behavior-bug-bounty-reporting-ai-safety-risksOpenAI launches a bug bounty program targeting model abuse and safety risks. Learn how to report jailbreaks and bypasses to improve enterprise AI security.Fri, 27 Mar 2026 16:25:44 GMTOpenAIBug BountyLLM SecurityAI SafetyPrompt Injectionhttps://runtimerebel.com/blog/claude-chrome-extension-zero-click-prompt-injection-vulnerabilityhttps://runtimerebel.com/blog/claude-chrome-extension-zero-click-prompt-injection-vulnerabilityA critical flaw in Anthropic's Claude Chrome extension allowed websites to silently inject malicious prompts using zero-click XSS techniques.Thu, 26 Mar 2026 16:31:26 GMTAnthropicClaudePrompt InjectionBrowser ExtensionXSShttps://runtimerebel.com/blog/claudy-day-prompt-injection-and-xss-flaws-target-claude-ai-usershttps://runtimerebel.com/blog/claudy-day-prompt-injection-and-xss-flaws-target-claude-ai-usersResearchers uncover 'Claudy Day', a trio of vulnerabilities in Anthropic's Claude AI that allow data theft through malicious Google search results.Wed, 18 Mar 2026 16:30:53 GMTAnthropicClaudePrompt InjectionData TheftLLM Securityhttps://runtimerebel.com/blog/hiding-malicious-commands-from-ai-via-font-rendering-manipulationhttps://runtimerebel.com/blog/hiding-malicious-commands-from-ai-via-font-rendering-manipulationLearn how attackers use font-rendering tricks to bypass AI safety filters and execute prompt injection attacks against LLM-powered assistants.Tue, 17 Mar 2026 16:30:27 GMTAI SecurityPrompt InjectionFont RenderingLLM VulnerabilitiesAdversarial AIhttps://runtimerebel.com/blog/openclaw-ai-agent-flaws-prompt-injection-and-data-exfiltration-riskhttps://runtimerebel.com/blog/openclaw-ai-agent-flaws-prompt-injection-and-data-exfiltration-riskCNCERT warns of critical security flaws in OpenClaw AI agents, enabling prompt injection and data exfiltration due to weak default configurations.Sat, 14 Mar 2026 20:09:05 GMTOpenClawAI SecurityPrompt InjectionCNCERTData Exfiltration