AI Alignment problems:
1. Capabilities generalize beyond alignment
2. System knowingly deceives
3. Does things you didn't want it to do
4. Limited or no interpretability of what's happening insideSounds like my mind.
AI Alignment problems:
1. Capabilities generalize beyond alignment
2. System knowingly deceives
3. Does things you didn't want it to do
4. Limited or no interpretability of what's happening insideSounds like my mind.