Overview
Northflank, a cloud platform for deploying applications, containers, databases and AI workloads, disclosed a security incident on October 30, 2026 and told customers to rotate every credential stored on the platform. Customers subsequently reported unauthorized use of AI API keys they had supplied to Northflank.
The disclosure followed a threat alert from cloud security firm Wiz at 11:43 UTC the same day. Northflank's guidance to customers was blunt: assume all environment variables have been decrypted, then disable and rotate all keys.
What We Know
- Disclosure: October 30, 2026, after a Wiz alert.
- Impact: Users reported unauthorized use of AI API keys stored on the platform.
- Containment: Northflank deployed a network policy patch, followed by further patches at 12:50 and 13:05 UTC, and brought in Wiz's incident response team.
- Added controls: Credential rotation was enabled for PostgreSQL, MongoDB, MySQL and Redis add-ons, and audit logs were enabled for all customers, including free-tier accounts.
What We Don't Know
Northflank has not said how the attackers got in, how many customers are affected, or whether credentials other than AI keys were stolen or misused. BleepingComputer's questions to the company had not been answered at the time of reporting.
Wider Context: "Black October" for AI Platforms
Wiz's report Black October for AI Developer Platforms counts at least nine incidents at AI development platforms during October, including Northflank, Railway, OpenCode, Warp and Lightning AI. In some of them, attackers went after environment variables holding AI credentials, a practice known as LLMJacking. In two cases Wiz investigated, stolen enterprise credentials drove roughly 2.5 million Amazon Bedrock API calls, about $100,000 in charges. Wiz has not confirmed that one actor is behind all nine incidents.
Recommended Actions
- Rotate all secrets stored in workloads, secret groups and global secrets on Northflank.
- Rotate database credentials for any add-ons, and disable public access to add-ons where it is enabled.
- Review usage and billing logs at every AI provider whose key was stored on the platform; set spend caps and alerts.
- Report suspicious usage to both Northflank and the affected AI provider.
- Treat any other secret that lived in an environment variable as exposed, and audit downstream systems for use of those secrets.
Takeaway
Environment variables are convenient but they concentrate high-value secrets in one place. Scope AI keys narrowly, set hard spend limits, and prefer short-lived credentials where a provider supports them.