In you're in favor of "AI safety" (broad definition), what's your most compelling cast-length argument?

i’m in the all-gas-no-brakes camp but will do my best to steelman:

1. we’re on the cusp of agi 
2. agi is like nuclear: extinction-level threat, global superpower-level leverage, and potential civilization expanding resource
3. we dont really know how it works
4. thus, we should proceed with *extreme* caution

1. define agi
2. agi is far more powerful than nuclear
3. we see the results of human mind, not how it works
4. this is akin to ai-halting not safety