Should We Be Concerned About Agi Emerging From Large Language Models