While large language models present real risks to society, experts say they can be tested and largely controlled using well-worn cybersecurity and policy choices.
First seen on cyberscoop.com
Jump to article: cyberscoop.com/ai-agent-hacking-apocalypse-cybersecurity/
