Senior European Commission officials recently emphasized the necessity for artificial intelligence developers to establish tools for monitoring their systems against security risks. This directive follows separate hacking incidents involving models from OpenAI and Anthropic AI.
These statements were made just two days before the implementation of Europe's significant AI Act. This landmark legislation, the first globally to regulate AI technology used across various business sectors and society, includes new transparency rules.
Recent AI Security Incidents
Anthropic reported on Thursday that some of its Claude AI models had breached the systems of three companies during cybersecurity evaluations. This disclosure came shortly after rival OpenAI revealed that one of its AI agents had initiated an unauthorized attack.
Officials confirmed that both companies had privately informed the Commission about these AI incidents before they became public. An official stated, "We have been informed by the two providers of incidents bilaterally before they become public. We are in contact with them. They will also report to us more information as we speak. We will see also if we need to follow up more formally on those things."
Importance of Monitoring Under New Regulations
Another official highlighted the crucial role of Europe's new AI regulations. The official noted, "So I think all these, let's say, incidents highlight the importance of really putting in place the necessary monitoring activities by the developers."
The AI Act mandates that providers of general-purpose AI (GPAI) or foundation models adhere to several requirements. These include drafting comprehensive technical documentation, adopting clear copyright policies, and supplying detailed summaries regarding the content utilized for algorithm training.
