Vulnerabilities in AI-assisted technology can snowball into national security risks; building safeguards and governance standards from day one can mitigate the risks from spreading across the network and causing large scale disruptions