It's a nice idea, but how do you know the agent is aligned with what it thinks the intent is?
or moreso, what happens at compact boundaries where the agent completely forgets the intent
alt Hacker News
or moreso, what happens at compact boundaries where the agent completely forgets the intent