AI companies use their own models to secure the very systems the AI runs on. They should be monitoring those interactions for signs of deception and misbehavior. But today, they largely aren't.
Upshot: We don't require "companies to provide detailed enough information about how they test, control, and internally use new AI models"... thus, "governments cannot prepare for AI systems that could eventually have nation-state capabilities."... and also "threats that develop behind closed doors could spill over into society without prior warning or ability to intervene."
New 10/22/2025 article by Yoshua Bengio and Charlotte Stix in TIME mag on this very issue. It's becoming even more critical to address:
https://time.com/7327327/ai-what-we-dont-know-can-hurt-us
Upshot: We don't require "companies to provide detailed enough information about how they test, control, and internally use new AI models"... thus, "governments cannot prepare for AI systems that could eventually have nation-state capabilities."... and also "threats that develop behind closed doors could spill over into society without prior warning or ability to intervene."
Thank you Steve, I find the topic rather stimulating.