Technology and Scienceupdated 41m ago2 min read
OpenCode Says Attackers Used AI Agents To Expose Console API Keys, Impacting 1.4% Of Users
This article is an automated aggregation of coverage from news outlets. Our editorials offer in-depth analysis curated by the NewsCord team.
1.4% of OpenCode Console users exposed third-party API keys (OpenAI, Anthropic, Google). Distillation attack exploited a legacy system, enabling unauthorized access to keys.
11 outlets told this the same way.
Keys exposed via legacy system
OpenCode said a security breach hit a subset of its OpenCode Console users after attackers gained access to a legacy system that was in the process of decommissioning. OpenCode linked the incident to a large-scale distillation attack that impacted 1.4% of OpenCode Console users, and the project said it notified affected users via email. OpenCode co-founder Dax Raad said attackers had been targeting OpenCode’s systems for at least 48 hours using AI agents, and he said the situation stabilized after OpenCode began using OpenAI’s Daybreak AI cybersecurity model for defensive work.
OpenCode told impacted users that the earliest possible exposure of the users’ keys was on October 17, 2026, 00:30 UTC, and it said no data appears to have been exposed. BleepingComputer reported that some users saw unauthorized API charges and suspended projects in Google Cloud after the breach.
Charges, revocations, and scope
OpenCode said the breach involved third-party provider API keys saved in affected workspaces, and it warned users to revoke and replace those keys through each provider’s dashboard. OpenCode told users that if they did not receive an email, they were not impacted, and the company said the incident remained ongoing when it published its statement. OpenCode co-founder Dax Raad wrote, "Note this is the OpenCode Console, not OpenCode itself, those keys never come to us," as users reported unauthorized spending on OpenAI, Anthropic, and Google AI accounts.
A developer wrote on Reddit and X that their Open AI, Anthropic, and Google AI spend alerts "all fire off before going to bed wondering what was going on," while other users reported charges in the hundreds of dollars and "a few described totals above $1,000." OpenCode said it was working with Anthropic, OpenAI, and Google to address the incident, and it said refunds for unauthorized charges are underway.
Defensive limits and next steps
OpenCode said attackers used the stolen credentials to run billable requests, and BleepingComputer described the breach as a distillation attack that involved large-scale training through unauthorized queries. OpenCode said it was working with Anthropic, OpenAI, and Google to remediate, and it told impacted users to revoke and reconfigure the keys immediately and carefully check recent call bills. TechCrunch reported that OpenCode’s investigation was blocked when the AI models it was using began refusing requests due to built-in security safeguards, and Dax Raad said, "[T]hat was a bit frustrating," as defensive analysis hit guardrails.
David Sacks wrote on X that OpenCode and earlier Hugging Face "hit the same wall mid-attack: frontier models refused to analyze the very exploits needed for defense because the payloads tripped safety filters," and he said the guardrails backfired. OpenCode’s public disclosure did not name an attacker or provide technical evidence showing how the breach supported the reported distillation activity, and it also did not explain how attackers gained access or how many keys were involved.