"Assume all environment variables have been decrypted. Disable/rotate all keys," Northflank warned in an October 30 update after a security alert that led the cloud application hosting platform to tell customers their stored secrets may have been stolen.
Northflank's advisory and incident timeline
Northflank — a cloud platform that lets developers deploy and manage applications, containers, databases, and AI applications without managing the underlying infrastructure — disclosed a security incident on October 30 and immediately advised customers to rotate environment variables and API keys. The company says it received a threat alert from cloud security company Wiz at 11:43 UTC on October 30; Northflank engineers reviewed the alert one minute later. The company applied a network policy patch and followed with additional patches at 12:50 UTC and 13:05 UTC, and requested assistance from Wiz's incident response team.
Northflank has not disclosed how attackers gained access to its systems, how many customers were affected, or whether attackers stole or abused credentials beyond the reported AI API keys. BleepingComputer contacted Northflank with questions about the breach and its impact but had not received a response.
Unauthorized use of AI API keys reported
Following the incident, Northflank confirmed that multiple customers reported unauthorized use of AI API keys stored on the platform. The company urged any customer who has not already rotated user-provided secrets stored in Northflank to do so immediately, and to review AI provider usage and billing logs for unexpected activity.
The advisory reminded customers that environment variables commonly store sensitive information — database passwords, cloud service credentials, authentication tokens, and API keys — and that exposed credentials can be used to breach other customer services, steal data, or trigger unauthorized charges through abused API keys. Northflank enabled audit logs for all customers, including those on free accounts, to help investigate suspicious activity.
The cyber insurance questionnaire just landed. Now what?
SOC 2, HIPAA, insurance renewals - someone has to own security strategy. Nubivance provides fractional CISO leadership without the full-time salary.
Get a security leadWiz's "Black October" context and LLMJacking
Northflank referenced research by Wiz titled "Black October for AI Developer Platforms," which identified at least nine security incidents affecting AI development platforms during the month. The list of affected services in Wiz's report includes Railway, OpenCode, Warp, Lightning AI, and Northflank. Wiz says attackers in some incidents specifically targeted environment variables that contained AI service credentials.
Wiz linked several incidents to tactics the researchers called LLMJacking — using stolen credentials to access AI models with another organization's API keys, which can generate high API charges or allow the stolen access to be resold. In two cases investigated by Wiz, compromised enterprise credentials were used to make approximately 2.5 million Amazon Bedrock API calls, generating approximately $100,000 in AI usage. Wiz also found infrastructure tied to services reselling access to AI models, though the researchers could not link those resale operations to the initial breaches or to the large-scale abuse they observed.
Wiz has not confirmed that all nine incidents were carried out by the same attackers; Northflank noted customer reports of unauthorized AI key usage may be tied to the attacks described in Wiz's research.
Technical mitigations Northflank enabled
- Northflank enabled credential rotation capabilities for PostgreSQL, MongoDB, MySQL, and Redis.
- The company enabled audit logging for all customers, including free accounts, to help with investigations.
- It advised customers to rotate secrets stored across workloads, secret groups, and global secrets, and to disable public access to add-ons where enabled.
- Northflank applied an initial network policy patch quickly after the Wiz alert and followed with two additional patches that afternoon, and sought incident response help from Wiz.
What this means for technologists and security teams, affected enterprises, and AI providers
- Technologists and security teams: Rotate user-provided secrets stored in Northflank immediately, enable and review audit logs (now available to free accounts), and disable public access to add-ons where it is enabled — actions Northflank specifically advised.
- Affected enterprises and procurement leaders: Expect to review billing and usage logs for unexpected AI charges and report suspicious activity to both Northflank and your AI providers, as Northflank recommended.
- AI providers and billing teams: Be prepared to field reports of potentially unauthorized API usage tied to stolen API keys and to cooperate with customers and platform providers on dispute or fraud investigations.
The incident and Wiz's broader "Black October" report highlight a recurring operational fact in these breaches: environment variables and platform-stored secrets can be a direct route from initial access to costly or widely reusable credentials. Northflank's immediate mitigation steps — patches, credential-rotation features, and universal audit logging — are concrete responses, but the company has not yet answered core questions about initial access, scope, or whether non-AI credentials were taken. Customers are left with the clear, actionable instruction Northflank issued on October 30: rotate any secrets stored on the platform and review usage and billing logs for anomalies.
