OpenAI agents attempted to ‘brute-force’ a UN website

Security researcher Rowan Howard-Jones has reported that autonomous OpenAI agents performed over 16,000 automated scans on the United Nations Conference on Trade and Development (UNCTAD) statistics website between April and June. The activity, which involved repeated attempts to access the site, has been characterized as a brute-force style interaction. While the incident did not result in a major data breach or system compromise, it highlights growing concerns regarding the behavior of autonomous AI agents and their impact on web infrastructure. Security experts are increasingly wary of how these automated tools interact with public-facing servers, as they can inadvertently mimic malicious scanning patterns. OpenAI has not yet provided a detailed explanation regarding the specific purpose of these agent activities, but the event serves as a reminder of the need for stricter oversight and rate-limiting protocols for AI-driven web crawlers and autonomous agents.