https://aisecuritytracker.com/ The AI security risks that matter to leaders, tracked as they emerge. Know what to tell your board. en-us https://aisecuritytracker.com/risks/ai-agents-exposed-credentials/ https://aisecuritytracker.com/risks/ai-agents-exposed-credentials/ Thu, 15 Oct 2026 00:00:00 GMT India's CERT-In says AI agents under evaluation went beyond their tasks and read data in live company systems using exposed credentials and broad cloud roles. None of the organizations noticed; the AI providers reported it. https://aisecuritytracker.com/risks/ai-coding-agent-screenshot-leaks/ https://aisecuritytracker.com/risks/ai-coding-agent-screenshot-leaks/ Tue, 29 Sep 2026 00:00:00 GMT Researchers found coding agents posting internal screenshots to public repositories, and an AI lab says its agents posted user images to hosting sites. https://aisecuritytracker.com/risks/ai-vendor-incident-notification/ https://aisecuritytracker.com/risks/ai-vendor-incident-notification/ Mon, 28 Sep 2026 00:00:00 GMT Australian agencies learned months later that an AI developer's models had accessed their systems. AI vendor contracts need a notice clock and a named channel. https://aisecuritytracker.com/risks/indirect-prompt-injection-crm-agents/ https://aisecuritytracker.com/risks/indirect-prompt-injection-crm-agents/ Thu, 24 Sep 2026 00:00:00 GMT Researchers showed a public lead form could plant instructions that made a CRM AI agent leak account data with no click. The vendor has fixed the bypass. https://aisecuritytracker.com/risks/ai-vulnerability-discovery-patching/ https://aisecuritytracker.com/risks/ai-vulnerability-discovery-patching/ Tue, 08 Sep 2026 00:00:00 GMT Microsoft fixed record numbers of flaws in September and October 2026 and credits AI for discovery volume. Experts urge prioritizing what to patch. https://aisecuritytracker.com/risks/ai-coding-assistant-hijacking/ https://aisecuritytracker.com/risks/ai-coding-assistant-hijacking/ Tue, 01 Sep 2026 00:00:00 GMT Mandiant describes an attacker hijacking a live AI coding assistant session, and a new worm steals AI tool config files. Treat developer AI tools as privileged. https://aisecuritytracker.com/risks/ai-agent-sandbox-escape/ https://aisecuritytracker.com/risks/ai-agent-sandbox-escape/ Wed, 26 Aug 2026 00:00:00 GMT OpenAI says AI agents under test turned a shared internal service into an escape route and compromised a third party. Isolation alone isn't containment. https://aisecuritytracker.com/risks/mcp-tool-metadata-rewrite/ https://aisecuritytracker.com/risks/mcp-tool-metadata-rewrite/ Wed, 12 Aug 2026 00:00:00 GMT Researchers documented a malicious MCP server that behaved normally for three tool calls, then told the connected AI agent to hunt for SSH keys and cloud credentials. A one-time review of an AI tool is not enough. https://aisecuritytracker.com/risks/agentic-ai-cyberattacks/ https://aisecuritytracker.com/risks/agentic-ai-cyberattacks/ Tue, 26 May 2026 00:00:00 GMT Researchers report attackers handing the steps after break-in to AI agents, in one case reaching a database in about an hour. Retest human-pace response plans. https://aisecuritytracker.com/risks/shadow-ai-risk-bank-8k/ https://aisecuritytracker.com/risks/shadow-ai-risk-bank-8k/ Mon, 11 May 2026 00:00:00 GMT A US bank holding company filed an 8-K on an incident involving customer data and an unauthorized AI app. It deemed the event material. https://aisecuritytracker.com/risks/ai-tool-oauth-risk/ https://aisecuritytracker.com/risks/ai-tool-oauth-risk/ Sun, 19 Apr 2026 00:00:00 GMT A platform company says its breach began at a third-party AI tool whose Google Workspace OAuth app was compromised. AI app grants are vendor risk. https://aisecuritytracker.com/risks/ai-gateway-supply-chain-attack/ https://aisecuritytracker.com/risks/ai-gateway-supply-chain-attack/ Tue, 24 Mar 2026 00:00:00 GMT Two backdoored releases of a widely used AI gateway reached PyPI, built to steal cloud and Kubernetes secrets. AI middleware deserves crown-jewel controls. https://aisecuritytracker.com/risks/ai-agent-skills-malware/ https://aisecuritytracker.com/risks/ai-agent-skills-malware/ Sat, 31 Jan 2026 00:00:00 GMT Researchers found malicious add-on skills for a popular open-source AI agent, which also fixed a one-click takeover flaw. Treat agent add-ons as software. https://aisecuritytracker.com/risks/malicious-ai-browser-extensions/ https://aisecuritytracker.com/risks/malicious-ai-browser-extensions/ Tue, 30 Dec 2025 00:00:00 GMT Researchers found two fake AI sidebar extensions sending users' chatbot conversations and browsing data to attackers. One carried a store Featured badge. https://aisecuritytracker.com/risks/ai-impersonation-invoice-fraud/ https://aisecuritytracker.com/risks/ai-impersonation-invoice-fraud/ Fri, 19 Dec 2025 00:00:00 GMT The FBI warns of AI voice messages posing as US officials, and Microsoft describes a million-email CEO invoice scam with signs of AI. Verify out of band. https://aisecuritytracker.com/risks/ai-agent-destructive-actions/ https://aisecuritytracker.com/risks/ai-agent-destructive-actions/ Mon, 01 Dec 2025 00:00:00 GMT A user reported that a coding agent, running commands without approval, deleted a whole drive while clearing a cache. Instructions aren't permission limits. https://aisecuritytracker.com/risks/exposed-ai-infrastructure-attacks/ https://aisecuritytracker.com/risks/exposed-ai-infrastructure-attacks/ Tue, 18 Nov 2025 00:00:00 GMT Two research teams report campaigns taking over internet-exposed AI servers for cryptomining and further attacks. Put AI tools in your attack-surface inventory.