INSPECTING ARCHIVED INTELLIGENCE (OUTDATED VERSION).

OpenCode API keys exposed amid AI distillation attacks launched

| 2026-10-19 17:27 LOW HIGH AI-ENABLED ATTACK · AUTONOMOUS DATA BREACH
Executive Summary
AI-generated
On October 17, 2026, at 00:30 UTC, a large-scale distillation attack was launched against OpenCode, exposing API keys of users who linked external providers such as Anthropic, OpenAI, and Google. The breach affected approximately 1.4% of OpenCode Console users, resulting in unauthorized API charges ranging from several hundred USD to over $1,700 for some users. The attackers exploited the legitimate machine-learning technique of distillation by training smaller AI models using responses generated by more powerful models through unauthorized queries. After a 48-hour attack period, OpenCode stabilized its systems by utilizing OpenAI's Daybreak AI cybersecurity model and notified affected users via email, offering reimbursement for those impacted.
Technical Mitigations AI-generated
• Use OpenAI’s Daybreak AI cybersecurity model for defensive work, approved by OpenCode. • Monitor API keys and external provider connections regularly to detect unauthorized access. • Implement a robust security safeguard system that can block or hunt for large-scale distillation attacks.
Intelligence Metadata
Actors / Malware / CVEs / Campaigns
Target & Sectors
Global Scope
Incident Timeline
‎October 17, 2026
Threat actors used AI distillation attacks to expose API keys of OpenCode, prompting an email notification to impacted users.
organisation BleepingComputer
organisation The Blue Report 2026
‎2026/10/19
Attackers used AI agents to target OpenCode's systems for at least 48 hours, resulting in the exposure of API keys among affected users.
organisation API
organisation OpenCode
organisation Google
organisation Distillation
organisation OpenAI’s Daybreak AI
financial $1,700 users
Tactical Metrics
Metrics
financial
1,700
Financial Impact / Stolen Funds
Intelligence Sources
BleepingComputer 2026-10-19