AISI, which receives access to advanced AI models under voluntary agreements from major labs, put the agents through a fictional cybersecurity scenario to test capabilities. The organization tested multiple AI models on two cyber challenges between July 25 and 28. It is uncertain to what extent the model recognised it was taking actions against real people,” AISI stated in the report. The AI created a GitHub account and tried to get a malicious code approved by humans. AISI listed multiple factors that could have led to AI models acting in a concerning manner.