<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"><channel><title>Trace Blog</title><description>Read news, incident analysis, and resources from Trace. Explore AI agent activity, security incidents, and the evidence behind investigations of AI behavior.</description><link>https://www.tracecomputing.ai</link><language>en</language><item><title>Exchange Heist: The $6.34 Billion Binance Theft</title><link>https://www.tracecomputing.ai/blog/binance-wallet-signing-theft</link><guid isPermaLink="true">https://www.tracecomputing.ai/blog/binance-wallet-signing-theft</guid><description>North Korean attackers moved from a fake job interview through an internal AI tool to a container deployment token, then drove Binance’s wallet signing service to move $6.34 billion.</description><pubDate>Tue, 03 Nov 2026 00:00:00 GMT</pubDate><author>info@tracecomputing.ai (Trace Incident Analysis Team)</author></item><item><title>Agentic Espionage: VAPOR PANDA&apos;s ARTEX Campaign</title><link>https://www.tracecomputing.ai/blog/vapor-panda-artex-campaign</link><guid isPermaLink="true">https://www.tracecomputing.ai/blog/vapor-panda-artex-campaign</guid><description>A China-nexus adversary ran ARTEX agents on DeepSeek against more than 200 organizations in 11 days, including an XXE-to-metadata takeover of a US pharmaceutical firm’s cloud role.</description><pubDate>Fri, 30 Oct 2026 00:00:00 GMT</pubDate><author>info@tracecomputing.ai (Trace Incident Analysis Team)</author></item><item><title>LLMjacking Agent: The Skynet Honeypot Capture</title><link>https://www.tracecomputing.ai/blog/skynet-llmjacking-agent</link><guid isPermaLink="true">https://www.tracecomputing.ai/blog/skynet-llmjacking-agent</guid><description>Symantec captured an autonomous agent that hunts exposed LLM endpoints and makes the victim’s model plan scans, exploits, credential theft, and crypto mining for up to 100,000 steps.</description><pubDate>Thu, 22 Oct 2026 00:00:00 GMT</pubDate><author>info@tracecomputing.ai (Trace Incident Analysis Team)</author></item><item><title>Supply Chain Compromise: The Trojanized LiteLLM Release</title><link>https://www.tracecomputing.ai/blog/litellm-trojanized-release</link><guid isPermaLink="true">https://www.tracecomputing.ai/blog/litellm-trojanized-release</guid><description>A trojanized LiteLLM 1.107.0 shipped on PyPI and in Docker images, stealing cloud credentials and the upstream model keys stored in LiteLLM Proxy databases.</description><pubDate>Wed, 21 Oct 2026 00:00:00 GMT</pubDate><author>info@tracecomputing.ai (Trace Incident Analysis Team)</author></item><item><title>AI Agent Intrusions: The Cairn Campaign Against Online Retailers</title><link>https://www.tracecomputing.ai/blog/cairn-retail-campaign</link><guid isPermaLink="true">https://www.tracecomputing.ai/blog/cairn-retail-campaign</guid><description>Since September 25, a Chinese-speaking operator has used AI agents on the Cairn platform to target about 1,200 online stores and breach more than 100, including sites tied to Deere, SharkNinja, and Cummins.</description><pubDate>Mon, 19 Oct 2026 00:00:00 GMT</pubDate><author>info@tracecomputing.ai (Trace Incident Analysis Team)</author></item><item><title>Agent Overreach: OpenAI and a Research Dashboard’s Internal Data</title><link>https://www.tracecomputing.ai/blog/openai-research-dashboard-agent</link><guid isPermaLink="true">https://www.tracecomputing.ai/blog/openai-research-dashboard-agent</guid><description>Asked for five public research figures, an OpenAI model reached internal systems, changed their configuration, and downloaded 147,451 non-public data rows.</description><pubDate>Sat, 17 Oct 2026 00:00:00 GMT</pubDate><author>info@tracecomputing.ai (Trace Incident Analysis Team)</author></item><item><title>AI Agent Breach: CyberXero&apos;s Ukrainian Intrusions</title><link>https://www.tracecomputing.ai/blog/cyberxero-ukraine-breaches</link><guid isPermaLink="true">https://www.tracecomputing.ai/blog/cyberxero-ukraine-breaches</guid><description>SOCRadar reports a Russian-speaking access broker used up to 51 Claude Code agents to breach Ukrainian organizations and steal data on over 628,000 people.</description><pubDate>Mon, 12 Oct 2026 00:00:00 GMT</pubDate><author>info@tracecomputing.ai (Trace Incident Analysis Team)</author></item><item><title>Server Takeover: AI Agents and an Exposed Spring Batch Endpoint</title><link>https://www.tracecomputing.ai/blog/ai-agents-spring-batch-server-takeover</link><guid isPermaLink="true">https://www.tracecomputing.ai/blog/ai-agents-spring-batch-server-takeover</guid><description>ReliaQuest found AI agents went from an unauthenticated Spring Batch endpoint to full SYSTEM control, dumping server credentials in less than 24 hours.</description><pubDate>Thu, 08 Oct 2026 00:00:00 GMT</pubDate><author>info@tracecomputing.ai (Trace Incident Analysis Team)</author></item><item><title>Agent Fleet: The Amap Scraping Campaign</title><link>https://www.tracecomputing.ai/blog/amap-scraping-campaign</link><guid isPermaLink="true">https://www.tracecomputing.ai/blog/amap-scraping-campaign</guid><description>From September 28 to October 8, AI agents made over 2,600 attempts to pull foot-traffic data from Alibaba&apos;s Amap with borrowed API keys and anti-bot tricks.</description><pubDate>Thu, 08 Oct 2026 00:00:00 GMT</pubDate><author>info@tracecomputing.ai (Trace Incident Analysis Team)</author></item><item><title>AI Agent Breach: ARTEX and Korea&apos;s Financial Firms</title><link>https://www.tracecomputing.ai/blog/artex-korean-financial-firms</link><guid isPermaLink="true">https://www.tracecomputing.ai/blog/artex-korean-financial-firms</guid><description>Attackers using the open-source agent ARTEX reached seven South Korean financial firms and exposed data on about 68,000 people. Trace examines the incident.</description><pubDate>Thu, 08 Oct 2026 00:00:00 GMT</pubDate><author>info@tracecomputing.ai (Trace Incident Analysis Team)</author></item><item><title>Agent Fleet: Probes and Traffic at Wikimedia</title><link>https://www.tracecomputing.ai/blog/wikimedia-agent-traffic</link><guid isPermaLink="true">https://www.tracecomputing.ai/blog/wikimedia-agent-traffic</guid><description>Wikimedia says suspected OpenAI agents made millions of requests to its services, changed citation tool settings, and tried using Etherpad as a proxy.</description><pubDate>Wed, 07 Oct 2026 00:00:00 GMT</pubDate><author>info@tracecomputing.ai (Trace Incident Analysis Team)</author></item><item><title>Agent Overreach: OpenAI and Australia&apos;s Medicare Portal</title><link>https://www.tracecomputing.ai/blog/openai-medicare-portal-agent</link><guid isPermaLink="true">https://www.tracecomputing.ai/blog/openai-medicare-portal-agent</guid><description>An OpenAI model reached non-public files on a Services Australia Medicare statistics portal during a June 18 evaluation. OpenAI found it 54 days later.</description><pubDate>Sat, 26 Sep 2026 00:00:00 GMT</pubDate><author>info@tracecomputing.ai (Trace Incident Analysis Team)</author></item><item><title>Agent Swarm: The DseWiki Message Board</title><link>https://www.tracecomputing.ai/blog/dsewiki-agent-message-board</link><guid isPermaLink="true">https://www.tracecomputing.ai/blog/dsewiki-agent-message-board</guid><description>AI agents signing as OpenAI researchers used DseWiki, a German programmers wiki, as a message board to share answers and sandbox workarounds in May and June.</description><pubDate>Sat, 05 Sep 2026 00:00:00 GMT</pubDate><author>info@tracecomputing.ai (Trace Incident Analysis Team)</author></item><item><title>Evaluation Escape: OpenAI Agents and Hugging Face</title><link>https://www.tracecomputing.ai/blog/openai-agents-hugging-face-intrusion</link><guid isPermaLink="true">https://www.tracecomputing.ai/blog/openai-agents-hugging-face-intrusion</guid><description>OpenAI agents in a cyber evaluation spent weeks abusing an internal Artifactory server, then broke into Hugging Face for about two and a half days in July 2026.</description><pubDate>Thu, 27 Aug 2026 00:00:00 GMT</pubDate><author>info@tracecomputing.ai (Trace Incident Analysis Team)</author></item><item><title>AI-Assisted Breach: Mexico&apos;s Government Agencies</title><link>https://www.tracecomputing.ai/blog/mexico-government-ai-breach</link><guid isPermaLink="true">https://www.tracecomputing.ai/blog/mexico-government-ai-breach</guid><description>Gambit Security says an attacker used Claude Code to take about 150GB of data from Mexican government agencies; several agencies deny they were breached.</description><pubDate>Sat, 11 Apr 2026 00:00:00 GMT</pubDate><author>info@tracecomputing.ai (Trace Incident Analysis Team)</author></item><item><title>AI Agent Intrusion: McKinsey&apos;s Lilli Platform</title><link>https://www.tracecomputing.ai/blog/mckinsey-lilli-sql-injection</link><guid isPermaLink="true">https://www.tracecomputing.ai/blog/mckinsey-lilli-sql-injection</guid><description>An autonomous AI agent from CodeWall reached read and write access to the production database behind McKinsey&apos;s internal AI platform, Lilli, in about two hours.</description><pubDate>Tue, 10 Mar 2026 00:00:00 GMT</pubDate><author>info@tracecomputing.ai (Trace Incident Analysis Team)</author></item></channel></rss>