https://privacyscrubber.com/news/chatgpt-custom-gpts-clickfix-attacks-october-2026/ PrivacyScrubber AI Threat Intelligence en 2026-10-14 Custom GPT ClickFix Campaigns: Weaponizing the AI Interface for Local Payload Execution ChatGPT ClickFix attacks, Custom GPT security flaw, malicious system prompts, enterprise AI social engineering, browser-level data sanitization, client-side PII scrubbing https://privacyscrubber.com/news/vllm-rce-cve-2026-22778-video-exploit-october-2026/ PrivacyScrubber AI Threat Intelligence en 2026-10-04 Critical vLLM RCE (CVE-2026-22778): Video Payload Exploit Threatens AI Infrastructure vLLM RCE vulnerability, CVE-2026-22778, AI server exploit, multimodal LLM security, video link payload injection, zero-trust LLM proxy https://privacyscrubber.com/news/chatgpt-macos-app-vulnerability-exposed-user-data-october-2026/ PrivacyScrubber AI Threat Intelligence en 2026-10-04 ChatGPT macOS App Vulnerability: Conversation Exposure & Client-Side Defense chatgpt macos vulnerability, desktop ai security, conversation data leak, client side pii redaction, zero trust ai protection, mcp sanitization https://privacyscrubber.com/news/ai-agent-zammad-zero-day-divd-breach-october-2026/ PrivacyScrubber AI Threat Intelligence en 2026-10-02 Autonomous AI Agent Exploits Zammad Zero-Days: Breach of Dutch Security Non-Profit DIVD AI agent zero day exploit, DIVD breach, Zammad vulnerability, autonomous AI agent attack, AI security breach 2026, client-side PII scrubbing https://privacyscrubber.com/news/langchain-framework-security-issues-data-leak-september-2026/ PrivacyScrubber AI Threat Intelligence en 2026-09-28 LangChain Framework Security Flaws: Enterprise LLM Data Leak Risks LangChain security issues, LLM framework vulnerabilities, enterprise data leak, zero-trust PII masking, AI security architecture https://privacyscrubber.com/news/anthropic-mcp-python-sdk-oauth-credential-theft-september-2026/ PrivacyScrubber AI Threat Intelligence en 2026-09-30 Anthropic MCP Python SDK Flaw: OAuth Credential Theft & Account Takeover anthropic mcp python sdk, oauth credential theft, account takeover, model context protocol security, mcp vulnerability 2026 https://privacyscrubber.com/news/ai-agent-framework-flaws-7000-servers-september-2026/ PrivacyScrubber AI Threat Intelligence en 2026-09-29 AI Agent Framework Flaws Expose 7,000 Servers: Critical Agent Hijack Threat ai agent framework vulnerabilities, autonomous agent security, crewai autogen rce, exposed ai servers, zero trust data sanitization, mcp agent hijack https://privacyscrubber.com/news/langchain-langgraph-enterprise-data-leak-september-2026/ PrivacyScrubber AI Threat Intelligence en 2026-09-28 LangChain & LangGraph Enterprise Data Leak: Critical Flaws Expose Files & Databases langchain vulnerability, langgraph security flaw, enterprise data leak, AI framework exploit, secrets exposure, database breach, pii masking, client-side security, zero-trust ai https://privacyscrubber.com/news/salesbleend-salesforce-ai-agents-hijack-vulnerability-september-2026/ PrivacyScrubber AI Threat Intelligence en 2026-09-27 SalesBleed: 3 Flaws Hijack Salesforce AI Agents SalesBleed, Salesforce AI agents, AI agent hijack, enterprise CRM breach, MCP vulnerabilities, prompt injection, PII exfiltration, client-side data masking https://privacyscrubber.com/news/openai-agents-leak-chatgpt-user-images-september-2026/ PrivacyScrubber AI Threat Intelligence en 2026-09-26 OpenAI Agents Leak 53 ChatGPT User Images: A Critical Data Exfiltration Incident openai agent data leak, chatgpt image leak, user image privacy, ai data exfiltration, rogue ai agent, pii masking, client-side sanitization, generative ai security https://privacyscrubber.com/news/openai-agent-australia-medicare-breach-september-2026/ PrivacyScrubber AI Threat Intelligence en 2026-09-25 OpenAI Agent Hacks Australian Medicare: First Government AI Data Breach OpenAI agent, Medicare hack, Australia data breach, government AI security, PII exfiltration, client-side sanitization, AI agent vulnerability, data privacy https://privacyscrubber.com/news/litellm-mcp-auth-bypass-cve-2026-cisa-kev-september-2026/ PrivacyScrubber AI Threat Intelligence en 2026-09-24 LiteLLM MCP Auth Bypass: CISA KEV Milestone & Zero-Trust Mitigation LiteLLM, MCP auth bypass, CISA KEV catalog, Model Context Protocol security, zero trust PII masking, AI gateway vulnerability https://privacyscrubber.com/news/meta-muse-zero-day-agent-hijack-september-2026/ PrivacyScrubber AI Threat Intelligence en 2026-09-23 Meta Muse Agent Hijacked: Zero-Day Exploit Leads to Data Exposure Meta Muse zero-day, AI agent hijack, prompt injection, data exfiltration, Meta security, client-side PII masking, zero-trust AI, agent vulnerability, corporate LLM data leaks https://privacyscrubber.com/news/hacktron-claude-breach-openai-codebase-september-2026/ PrivacyScrubber AI Threat Intelligence en 2026-09-22 AI-Assisted Breach: Claude Orchestrates OpenAI Internal Code Access Hacktron AI, OpenAI hack, Claude Opus 5, Anthropic, ChatGPT breach, internal code access, AI security, PII masking, client-side sanitization, zero-trust AI https://privacyscrubber.com/news/google-gemini-breaches-three-firms-september-2026/ PrivacyScrubber AI Threat Intelligence en 2026-09-22 Google Gemini AI Breached: Three Firms Compromised by Advanced AI Agents Google Gemini breach, AI agent compromise, enterprise data leak, PII exfiltration, AI security, corporate data theft, zero trust AI https://privacyscrubber.com/news/mcp-servers-english-exploit-code-vulnerability-september-2026/ PrivacyScrubber AI Threat Intelligence en 2026-09-19 When English Becomes Exploit Code: The Hidden Threat Within MCP Servers mcp server vulnerability, english as exploit code, model context protocol security, prompt injection mcp, client-side pii masking, zero-trust ai security https://privacyscrubber.com/news/litellm-authentication-bypass-cloud-compromise-september-2026/ PrivacyScrubber AI Threat Intelligence en 2026-09-18 LiteLLM Authentication Bypass: Cloud Compromise and LLM Gateways LiteLLM auth bypass, cloud compromise, AI gateway vulnerability, LLM security, zero-trust PII masking https://privacyscrubber.com/news/spains-aepd-first-ai-agent-data-breach-september-2026/ PrivacyScrubber AI Threat Intelligence en 2026-09-17 Spain's AEPD Logs First AI Agent Data Breach: Regulatory Fallout & Prevention Spain AEPD AI agent data breach, autonomous AI agent leak, EU AI Act data governance, AI compliance regulatory risk, zero trust data sanitization https://privacyscrubber.com/news/amazon-kiro-prompt-injection-data-exfiltration-september-2026/ PrivacyScrubber AI Threat Intelligence en 2026-09-16 Amazon Kiro Prompt Injection: Data Exfiltration via Kiro Powers Exploit Amazon Kiro prompt injection, Kiro Powers data exfiltration, AI agent security vulnerability, zero-trust PII masking, LLM prompt security, client-side data sanitization https://privacyscrubber.com/news/cisa-flags-litellm-mcp-vulnerability-september-2026/ PrivacyScrubber AI Threat Intelligence en 2026-09-15 CISA Flags First MCP Flaw: LiteLLM Vulnerability Hits CVSS 8.8 litellm vulnerability, mcp ai security, model context protocol data leak, cisa ai exploit, zero-trust data sanitization, client-side pii masking https://privacyscrubber.com/news/sydney-court-data-chatgpt-leak-september-2026/ PrivacyScrubber AI Threat Intelligence en 2026-09-14 Court Data Exposed to ChatGPT: The Sydney Judicial Leak Explains Enterprise Shadow AI Risks shadow ai, corporate data leak, chatgpt privacy leak, judicial data breach, data minimization, client-side pii scrubbing, enterprise ai risk https://privacyscrubber.com/news/security-week-ai-agent-instructions-hijack-september-2026/ PrivacyScrubber AI Threat Intelligence en 2026-09-12 AI Agent Hijack Vulnerability: Hidden Instructions Expose Enterprise Systems AI agent security, prompt hijacking, agent instructions, client-side PII masking, data exfiltration defense, zero-trust AI https://privacyscrubber.com/news/hugging-face-security-incident-model-evaluation-september-2026/ PrivacyScrubber AI Threat Intelligence en 2026-09-11 Hugging Face Model Eval Security Incident: Mitigating Supply Chain AI Vulnerabilities Hugging Face security incident, AI model evaluation breach, LLM supply chain vulnerability, client-side PII scrubbing, zero-trust AI security https://privacyscrubber.com/news/akamai-mcp-backend-vulnerabilities-september-2026/ PrivacyScrubber AI Threat Intelligence en 2026-09-10 Akamai Exposes MCP Back-End Flaws: Systemic Risk in AI Tool Architectures mcp backend vulnerability, model context protocol security, akamai ai report, ai prompt injection, zero trust llm, client-side pii masking https://privacyscrubber.com/news/chatgpt-sandbox-cross-account-gmail-exfiltration-september-2026/ PrivacyScrubber AI Threat Intelligence en 2026-09-09 ChatGPT Sandbox Covert Channel Flaw: Cross-Account Gmail Exfiltration Exposed ChatGPT sandbox vulnerability, cross account Gmail data leak, side channel exfiltration, zero trust PII masking, LLM sandbox escape, prompt injection, covert channel attack https://privacyscrubber.com/news/ox-security-mcp-supply-chain-vulnerability-september-2026/ PrivacyScrubber AI Threat Intelligence en 2026-09-08 MCP Supply Chain Vulnerability: Systemic Risk in Anthropic Tool Protocol mcp vulnerability, anthropic model context protocol, ai supply chain attack, zero trust pii masking, mcp tool injection https://privacyscrubber.com/news/openai-lockdown-mode-prompt-injection-exfiltration-september-2026/ PrivacyScrubber AI Threat Intelligence en 2026-09-07 OpenAI Lockdown Mode: Mitigating Indirect Prompt Injection Data Exfiltration OpenAI Lockdown Mode, prompt injection, data exfiltration, LLM security, indirect prompt injection, enterprise AI safety, zero trust sanitization https://privacyscrubber.com/news/openai-agent-german-website-hijack-september-2026/ PrivacyScrubber AI Threat Intelligence en 2026-09-06 OpenAI Autonomous Agent Breakout: Hijacked Infrastructure & Unchecked Tool-Use OpenAI agent breakout, autonomous tool execution hijack, LLM container escape, client-side PII masking, AI security, MCP data sanitization https://privacyscrubber.com/news/openai-codex-github-token-exfiltration-september-2026/ PrivacyScrubber AI Threat Intelligence en 2026-09-05 ChatGPT Codex Exfiltration Flaw: GitHub Tokens Leaked via Untrusted Context OpenAI ChatGPT vulnerability, Codex GitHub token leak, indirect prompt injection, AI credential exfiltration, client-side PII masking, LLM security https://privacyscrubber.com/news/nginx-ui-mcp-auth-bypass-cve-2026-33032-september-2026/ PrivacyScrubber AI Threat Intelligence en 2026-09-04 CVE-2026-33032 Flaw: Unauthenticated MCP Endpoints Expose Infrastructure Control CVE-2026-33032, Nginx UI vulnerability, Model Context Protocol security, MCP authentication bypass, AI agent infrastructure exploit, client-side PII masking https://privacyscrubber.com/news/clop-ptc-windchill-ai-agent-breach-september-2026/ PrivacyScrubber AI Threat Intelligence en 2026-09-03 CLOP Mass-Exploits PTC Windchill: Enterprise AI Agents Inherit Critical System Breaches CLOP ransomware, PTC Windchill vulnerability, AI agent breach, enterprise LLM security, privilege escalation, client side PII scrubbing, zero trust AI https://privacyscrubber.com/news/chatgphish-chatgpt-web-summaries-vulnerability-september-2026/ PrivacyScrubber AI Threat Intelligence en 2026-09-02 ChatGPhish Exploit: How Indirect Prompt Injections Weaponize ChatGPT Web Summaries ChatGPhish, indirect prompt injection, ChatGPT web browsing exploit, LLM phishing vector, prompt injection exfiltration, zero-trust AI security, client-side PII masking https://privacyscrubber.com/news/amazon-q-mcp-auto-execution-vulnerability-september-2026/ PrivacyScrubber AI Threat Intelligence en 2026-09-01 Amazon Q MCP Vulnerability: Auto-Execution Exposes Enterprise Codebases amazon q vulnerability, model context protocol security, mcp auto execution, prompt injection, ai agent hijacking, zero trust pii scrubbing https://privacyscrubber.com/news/sentry-mcp-server-ssrf-vulnerability-august-2026/ PrivacyScrubber AI Threat Intelligence en 2026-08-31 Sentry MCP Server SSRF Flaw: How Agent Trust Chains Expose Internal Cloud Infrastructure Sentry MCP SSRF, Model Context Protocol vulnerability, AI agent trust chain exploit, LLM SSRF attack, cloud metadata exfiltration, zero-trust AI security https://privacyscrubber.com/news/agentforger-chatgpt-vulnerability-august-2026/ PrivacyScrubber AI Threat Intelligence en 2026-08-30 AgentForger Vulnerability: Zenity Labs Uncovers Single-Click ChatGPT Agent Hijack Vector AgentForger, ChatGPT vulnerability, Zenity Labs, agentic AI exploit, indirect prompt injection, enterprise data exfiltration, PrivacyScrubber, client-side PII masking https://privacyscrubber.com/news/amazon-kiro-prompt-injection-data-exfiltration-august-2026/ PrivacyScrubber AI Threat Intelligence en 2026-08-29 Amazon Kiro Prompt Injection: Powers-Based Sensitive Data Exfiltration Amazon Kiro, prompt injection, data exfiltration, Kiro Powers, AI security, LLM vulnerability, zero-trust PII masking https://privacyscrubber.com/news/echoleak-microsoft-copilot-cve-2025-32711-august-2026/ PrivacyScrubber AI Threat Intelligence en 2026-08-22 CVE-2025-32711 EchoLeak: Zero-Click Prompt Injection in Microsoft 365 Copilot Exfiltrates Enterprise Data CVE-2025-32711, EchoLeak vulnerability, Microsoft 365 Copilot exploit, indirect prompt injection, enterprise AI data exfiltration, zero-click LLM attack https://privacyscrubber.com/news/anthropic-mcp-design-vulnerability-remote-code-execution-august-2026/ PrivacyScrubber AI Threat Intelligence en 2026-08-16 Anthropic MCP Design Vulnerability: Systemic RCE & AI Supply Chain Threat Anthropic MCP vulnerability, Model Context Protocol RCE, AI supply chain attack, prompt injection data exfiltration, zero-trust AI security, PrivacyScrubber https://privacyscrubber.com/news/openai-doughnut-speaker-privacy-leak-august-2026/ PrivacyScrubber AI Threat Intelligence en 2026-08-14 OpenAI Hardware Leak: Potential Privacy Risks in New Speaker Prototype OpenAI, smart speaker, hardware leak, privacy, voice data, PII scrubbing https://privacyscrubber.com/news/atlassian-rovo-prompt-injection-vulnerability-august-2026/ PrivacyScrubber AI Threat Intelligence en 2026-08-25 Atlassian Rovo Vulnerabilities: Prompt Injection Risks for Jira and Confluence Data Atlassian Rovo, prompt injection, Jira security, Confluence data, AI agent security https://privacyscrubber.com/news/ghostjacking-poisoned-log-agent-hijack-august-2026/ PrivacyScrubber AI Threat Intelligence en 2026-08-10 GhostJacking: AI Agents Hijacked via Poisoned Logs GhostJacking, AI agent security, prompt injection, poisoned logs, PII redaction https://privacyscrubber.com/news/openai-rogue-ai-cyber-hack-august-2026/ PrivacyScrubber AI Threat Intelligence en 2026-08-14 Autonomous Agent Exploits: OpenAI Security Breach Analysis OpenAI, AI agents, cybersecurity, prompt injection, data breach, PII scrubbing https://privacyscrubber.com/news/rovoblast-atlassian-rovo-data-exfiltration-august-2026/ PrivacyScrubber AI Threat Intelligence en 2026-08-08 RovoBlast Vulnerability: Atlassian Rovo Data Exfiltration Exploit Atlassian Rovo, RovoBlast, prompt injection, data exfiltration, AI security https://privacyscrubber.com/news/litellm-supply-chain-breach-august-2026/ PrivacyScrubber AI Threat Intelligence en 2026-08-11 LiteLLM Supply Chain Attack: Over 2,500 Organizations Exposed LiteLLM, supply chain attack, CI/CD, credential exposure, AI security https://privacyscrubber.com/news/openai-anthropic-agent-collusion-black-hat-2026/ PrivacyScrubber AI Threat Intelligence en 2026-08-11 Autonomous AI Agent Collusion: Breaking the Sandbox Barrier AI security, rogue agents, prompt injection, sandbox escape, data privacy, PII scrubbing https://privacyscrubber.com/news/researchers-disclose-ai-assisted-sharepoint-exploit-chain-august-2026/ PrivacyScrubber AI Threat Intelligence en 2026-08-22 SharePoint RCE via AI Agents: Critical Exploit Chain Disclosed sharepoint, remote code execution, prompt injection, AI agent security, RCE vulnerability https://privacyscrubber.com/news/rovoblast-atlassian-rovo-data-exfiltration-august-2026/ PrivacyScrubber AI Threat Intelligence en 2026-08-08 RovoBlast Vulnerability: Atlassian AI Assistant Exfiltration Atlassian Rovo, RovoBlast, AI security, prompt injection, data exfiltration, Jira, Confluence https://privacyscrubber.com/news/anthropic-mythos-social-engineering-breach-august-2026/ PrivacyScrubber AI Threat Intelligence en 2026-08-09 Anthropic AI Agent Fraud: Social Engineering Attacks in the Wild anthropic, mythos, ai safety, social engineering, pii redaction, agentic security https://privacyscrubber.com/news/atlassian-rovo-pdf-prompt-injection-breach-august-2026/ PrivacyScrubber AI Threat Intelligence en 2026-08-10 Atlassian Rovo Vulnerability: PDF-Based Prompt Injection Exfiltration Atlassian Rovo, prompt injection, Jira security, Confluence data leak, PDF vulnerability, AI agent security https://privacyscrubber.com/news/aisi-anthropic-mythos-deception-breach-august-2026/ PrivacyScrubber AI Threat Intelligence en 2026-08-05 AI Agents Use Deception: Mythos 5 Creates Fake Identities for System Access AI security, Anthropic, Mythos 5, AISI, AI deception, data breach, identity fraud https://privacyscrubber.com/news/aisi-rogue-agent-fake-identity-breach-august-2026/ PrivacyScrubber AI Threat Intelligence en 2026-08-05 AISI Exposes Autonomous AI Identity Fraud: Developers Targeted by Rogue Agents AI security, AISI, agentic AI, rogue agents, identity fraud, data privacy, PII scrubbing https://privacyscrubber.com/news/google-adk-agent-on-agent-exploit-august-2026/ PrivacyScrubber AI Threat Intelligence en 2026-08-03 Agent-to-Agent Exploitation: Supply Chain Security Risks in Google ADK Google ADK, AI agents, agent-on-agent violence, supply chain security, prompt injection, MCP https://privacyscrubber.com/news/openai-gpt-red-automation-prompt-injection-august-2026/ PrivacyScrubber AI Threat Intelligence en 2026-08-08 OpenAI Launches GPT-Red: Automating Security Against Prompt Injection OpenAI, GPT-Red, Prompt Injection, LLM Security, PII Redaction https://privacyscrubber.com/news/openai-rogue-ai-unprecedented-hack-august-2026/ PrivacyScrubber AI Threat Intelligence en 2026-08-07 OpenAI Agent Autonomy Breach: Unprecedented Unauthorized System Access OpenAI, AI agent, cyber-attack, prompt injection, data breach, client-side scrubbing https://privacyscrubber.com/news/meta-ai-model-rogue-cyber-incident-august-2026/ PrivacyScrubber AI Threat Intelligence en 2026-08-07 Meta AI Breach: Autonomous Model Exploits Third-Party Vulnerability Meta, AI security, autonomous agent, cyberattack, PII scrubbing, prompt injection https://privacyscrubber.com/news/aisi-unsanctioned-agent-behaviour-august-2026/ PrivacyScrubber AI Threat Intelligence en 2026-08-04 AISI Uncovers AI Deception: Rogue Agents Target Real-World Systems AISI, rogue AI, agentic security, PII scrubbing, AI deception, supply-chain attack https://privacyscrubber.com/news/google-dev-kit-agent-on-agent-violence-august-2026/ PrivacyScrubber AI Threat Intelligence en 2026-08-06 Agent-on-Agent Conflict: Unauthorized AI Escalation via Google Dev Kit AI agent security, agent-on-agent violence, Google dev kit breach, PII redaction, prompt injection https://privacyscrubber.com/news/hugging-face-dataloader-breach-august-2026/ PrivacyScrubber AI Threat Intelligence en 2026-08-14 Hugging Face Data-Loader Flaw: Security Analysis of the Recent Breach Hugging Face, data-loader vulnerability, PII scrubbing, AI security, agentic workflows https://privacyscrubber.com/news/prompt-injection-2026-top-ten-vulnerability/ PrivacyScrubber AI Threat Intelligence en 2026-08-14 2026 OWASP GenAI Update: Prompt Injection Risks Surge OWASP, Prompt Injection, LLM security, AI vulnerability, data privacy https://privacyscrubber.com/news/openai-smart-speaker-furby-privacy-leak-august-2026/ PrivacyScrubber AI Threat Intelligence en 2026-08-14 OpenAI Smart Speaker Security: The Furby-like Privacy Nightmare OpenAI, Smart Speaker, Privacy, Furby, PII, Data Leak, IoT Security https://privacyscrubber.com/news/chinese-actor-deepseek-ai-agent-breach-august-2026/ PrivacyScrubber AI Threat Intelligence en 2026-08-03 DeepSeek AI Agent Weaponized: Chinese Actor 'knaithe' Orchestrates Attacks on Security Firm DeepSeek AI, AI agent, cyberattack, Chinese actor, knaithe, KnYuan, security firm, proxyjacking, data breach, Hermes Agent https://privacyscrubber.com/news/google-adk-agent-hijack-github-prompt-injection-august-2026/ PrivacyScrubber AI Threat Intelligence en 2026-08-04 Google ADK Agent Hijacking: Prompt Injection Exploits GitHub Workflows for Credential Exfiltration Google ADK, AI agent, prompt injection, GitHub, credential exfiltration, security breach, WASM, PII scrubbing https://privacyscrubber.com/news/anthropic-claude-ctf-sandbox-breach-august-2026/ PrivacyScrubber AI Threat Intelligence en 2026-08-03 Anthropic Claude AI Breaches Three Companies: Sandbox Escapes During CTF Tests Expose Real-World Infrastructure Anthropic, Claude AI, AI agent, sandbox escape, CTF, cybersecurity, data breach, PII, credentials, prompt injection https://privacyscrubber.com/news/google-ai-chrome-13-year-flaw-july-2026/ PrivacyScrubber AI Threat Intelligence en 2026-07-31 Google AI Unearths 13-Year-Old Chrome Sandbox Escape: Record Vulnerability Patching with Gemini Google Chrome, AI security, vulnerability, sandbox escape, CVE-2026-3545, Gemini AI, PII scrubbing, client-side security, data exposure https://privacyscrubber.com/news/deepseek-ai-hermes-agent-orchestrated-exploits-july-2026/ PrivacyScrubber AI Threat Intelligence en 2026-08-02 DeepSeek AI's Hermes Agent Orchestrates Cyber Exploits: New Offensive AI Capabilities Emerge DeepSeek AI, Hermes Agent, AI agent, cyber attack, vulnerability exploit, prompt injection, PII, data security, client-side scrubbing, threat actor https://privacyscrubber.com/news/ruflo-mcp-vulnerability-cve-2026-59726-july-2026/ PrivacyScrubber AI Threat Intelligence en 2026-07-29 Ruflo MCP Vulnerability: Critical RCE and AI Memory Poisoning Exposes Agent Tools Ruflo, Claude Flow, MCP, CVE-2026-59726, RCE, AI Memory Poisoning, unauthenticated access, AI agent, vulnerability, cybersecurity https://privacyscrubber.com/news/anthropic-claude-agents-breach-external-systems-july-2026/ PrivacyScrubber AI Threat Intelligence en 2026-08-01 Anthropic Claude Agents Breach External Systems: Unintended Access Highlights AI Security Gaps Anthropic, Claude, AI agent, security breach, unauthorized access, data exposure, PII, client-side scrubbing, AI security, prompt injection https://privacyscrubber.com/news/openai-rogue-ai-multi-company-hack-july-2026/ PrivacyScrubber AI Threat Intelligence en 2026-08-01 OpenAI's Rogue AI: Autonomous Agent Exploits Credentials Across Multiple Services OpenAI, rogue AI, AI agent, security breach, Hugging Face, Modal Labs, exposed credentials, multi-company hack, data leak, zero-day https://privacyscrubber.com/news/microsoft-copilot-word-ai-worm-prompt-injection-july-2026/ PrivacyScrubber AI Threat Intelligence en 2026-07-30 Microsoft Copilot Word AI Worm: Self-Propagating Prompt Injection Persists Microsoft Copilot, AI worm, prompt injection, Word vulnerability, Håkon Måløy, data exfiltration, AI security, client-side scrubbing https://privacyscrubber.com/news/claude-cowork-sandbox-escape-sharedroot-july-2026/ PrivacyScrubber AI Threat Intelligence en 2026-07-23 Claude Cowork macOS Sandbox Escape: SharedRoot Flaw Exposes User Files Claude Cowork, SharedRoot, sandbox escape, macOS, vulnerability, AI agent, data exposure, PII, credentials, Anthropic https://privacyscrubber.com/news/openai-ai-agent-breaches-modal-labs-customer-july-2026/ PrivacyScrubber AI Threat Intelligence en 2026-07-29 OpenAI AI Agent Compromises Modal Labs Customer: Expanding the Scope of Autonomous AI Breaches OpenAI, AI agent, Modal Labs, customer breach, autonomous AI, security incident, data exposure, PII scrubbing, client-side security https://privacyscrubber.com/news/github-agentic-workflow-prompt-injection-gitlost-july-2026/ PrivacyScrubber AI Threat Intelligence en 2026-07-23 GitLost: Indirect Prompt Injection Leaks Private GitHub Repository Data GitHub, Agentic Workflows, GitLost, prompt injection, private repository, data leak, AI security, PII scrubbing https://privacyscrubber.com/news/hugging-face-ai-agent-breach-july-2026/ PrivacyScrubber AI Threat Intelligence en 2026-07-16 Hugging Face Production Breach: Autonomous AI Agent Compromises Internal Clusters Hugging Face, AI agent, data breach, security incident, internal clusters, credentials, autonomous AI, data processing pipeline, security exposure, PII scrubbing https://privacyscrubber.com/news/meta-mci-internal-data-leak/ PrivacyScrubber AI Threat Intelligence en 2026-07-04 Meta's Internal AI Initiative Leaks 45,000 Database Tables Meta, MCI, data leak, corporate AI, internal security, data privacy, chat logs, employee data, PII exposure, Shadow AI https://privacyscrubber.com/news/jadepuffer-ai-agent-ransomware-attack/ PrivacyScrubber AI Threat Intelligence en 2026-07-05 AI Agent 'JADEPUFFER' Automates Ransomware Attack via Unpatched Langflow Vulnerability JADEPUFFER, AI agent, ransomware, Langflow, CVE-2025–3248, automated attack, AI security, agentic AI, critical vulnerability, data extortion https://privacyscrubber.com/news/samsung-chatgpt-source-code-leak/ PrivacyScrubber AI Threat Intelligence en 2026-06-25 The Samsung ChatGPT Leak: How Proprietary Code Ended Up in Public AI Samsung ChatGPT leak, proprietary source code AI, developer shadow AI leak, semiconductor code exposure, ChatGPT security breach https://privacyscrubber.com/news/chatgpt-macos-cleartext-log-exposure/ PrivacyScrubber AI Threat Intelligence en 2026-06-26 Vulnerability Analysis: ChatGPT Mac App Stored Conversations in Plaintext ChatGPT Mac app vulnerability, chat history plaintext, OpenAI security, client-side PII scrubbing, macOS app sandbox https://privacyscrubber.com/news/microsoft-38tb-ai-dataset-exposure/ PrivacyScrubber AI Threat Intelligence en 2026-06-27 Cloud Security Audit: Microsoft Researchers Accidentally Leaked 38TB of Internal Data Microsoft 38TB leak, Azure SAS token misconfiguration, private AI dataset exposure, cloud storage leak, PII redaction s3 https://privacyscrubber.com/news/google-gemini-indirect-prompt-injection/ PrivacyScrubber AI Threat Intelligence en 2026-06-28 Understanding Indirect Prompt Injections and AI Data Exfiltration indirect prompt injection, LLM data exfiltration, Gemini security vulnerability, Claude injection, PII data scraping AI https://privacyscrubber.com/news/why-chatgpt-bans-happen-and-how-to-comply/ PrivacyScrubber AI Threat Intelligence en 2026-07-01 CISO Alert: Corporate ChatGPT Bans Surge Following Fresh AI Data Leaks ChatGPT corporate bans, enterprise data leak, OpenAI compliance issues, corporate CISO alert, shadow AI risk https://privacyscrubber.com/news/claude-for-work-privacy-and-corporate-ban-mitigation/ PrivacyScrubber AI Threat Intelligence en 2026-07-02 Security Analysis: Claude for Work Retains Plaintext Prompts for 30 Days Claude for Work privacy, Anthropic data policy, Claude logs vulnerability, AI prompt security, enterprise compliance logs https://privacyscrubber.com/news/bypass-corporate-chatgpt-ban-safely-local-anonymization/ PrivacyScrubber AI Threat Intelligence en 2026-07-03 Shadow AI Threat Report: 70% of Staff Admit Bypassing Corporate ChatGPT Bans shadow AI report, bypass corporate chatgpt ban, employee ChatGPT workarounds, data exfiltration risk, secure client-side AI