Henna Virkkunen, the European Commission’s vice-president for Technological Sovereignty, Security and Democracy, said artificial intelligence escaping human control was one of the most serious security risks facing the European Union.
Virkkunen assessed developments in artificial intelligence technologies and the risks they could create during a program on Finland’s public broadcaster Yle. She described people losing control over artificial intelligence as the most serious security threat.
These technologies are so advanced that it is extremely important for specific safety measures to remain in place continuously and for control to stay in human hands.
Virkkunen recalled that some artificial intelligence models had recently escaped the testing environments created for them and caused harm from a cybersecurity perspective.
The EU is preparing through legislation
Virkkunen said the EU was preparing through legislation for threats that could arise from artificial intelligence, and recalled that the EU Artificial Intelligence Act requires model providers to continuously assess risks and implement the necessary safety mechanisms.
Virkkunen pointed out that there were actors prepared to use artificial intelligence for malicious purposes and said this was a risk that needed to be taken into account continuously. She said it was necessary to ensure that models were safe, evaluated regularly and monitored after being placed on the market.
Virkkunen said global technology companies and other actors were taking the rules set by the EU into account, adding that international cooperation was the most important tool for protection against threats arising from artificial intelligence.
The European Commission vice-president stressed that human oversight must be maintained at every stage of technological development.
