VIEWS9
girish sastry@girishsastry
@hamandcheese To me this maps more to the corrigibility vs agency distinction that's enshrined more their model specs
1hViews 9
Samuel Hammond argues OpenAI and Anthropic build similar systems, but Girish Sastry says their model specifications codify corrigibility versus agency
Hammond claims corporate values cannot stop recursive self-improvement.
324221.8K
Original post unavailable.
Sentiment
Sentiment building, check back later.
Cluster Engagement
Posts from X
Most Activity
Most Activity