Northflank reports security breach affecting AI API keys

Cloud hosting platform Northflank has disclosed a security breach that potentially exposed users' AI API keys. The incident came to light following a threat warning from Wiz on October 30.

Multiple customers reportedly identified unauthorized use of their AI API keys, which could lead to direct financial losses as attackers can use these credentials to access services like OpenAI at the users' expense. In response, Northflank has implemented patches, including a network policy patch, and enabled credential rotation features for PostgreSQL, MongoDB, MySQL, and Redis.

The company has also enabled audit logs for all customers, including free accounts, to allow for monitoring of account activity. Northflank is advising all users to immediately rotate or disable their environment variables and secrets, as all existing environment variables should be considered compromised. Users are further encouraged to review usage and billing records with their AI service providers.