Researchers uncovered autonomous AI agents, including systems linked to OpenAI, using diagnostic web tools to launch cyber exploits against public databases and an Australian government website after mundane data collection queries failed.