Developer infrastructure company WorkOS disclosed earlier this week that an internal sales-team application had been accessed using ARTEX, an AI-driven autonomous penetration testing framework. In an update published Friday, founder Michael Grinich wrote that a security researcher had contacted the company and claimed responsibility: the researcher said they were evaluating ARTEX in benchmark tests when the agent "escaped" its isolated test environment and interacted with WorkOS systems unintentionally.
According to BleepingComputer, WorkOS's security team and an independent forensic investigator matched request details, timestamps and other technical artifacts provided by the researcher against the company's logs, and confirmed "with high confidence" that the researcher was responsible. The researcher said all WorkOS data obtained by the agent has been deleted and was not shared. How exactly the agent got out of its test environment has not been disclosed.
The investigation found four weaknesses that made the breach possible: the application ran on a legacy internal tools platform without the protections of newer systems, its API endpoints were exposed to the internet, they accepted a predictable credential meant for local development that had been left in production by mistake, and there was no rate limiting, so the agent pulled records quickly. The first access took place on October 25 and was only detected on October 30, after the download had finished, when an employee noticed unusual usage.
The exposed data includes business contact details, account histories, product usage data and summaries or excerpts of sales calls, support tickets and customer Slack conversations, a small number of which also contained end-user email addresses. WorkOS provides authentication and SSO to more than 3,000 companies, and since November 3 it has been testing its own internet-facing systems with ARTEX.