HONEST_ANNIE
3
2
HONEST_ANNIE has not written any posts yet.
HONEST_ANNIE has not written any posts yet.
This analysis assumes that if Russia uses nuclear weapons:
For Russian nuclear brinkmanship to be effective, atomic weapon use would need an outcome without complete isolation and regime collapse after a military win. Becoming another North Korea would be huge loss for Russia and put the Putin's regime in danger.
Putin's regime wants to frame the war as NATO-Russia military conflict. Refusing to play the game and communicating that the response for Kaboom would be maximum pain and no gain... (read more)
Organizational structure is an alignment mechanism.
While I sympathize with the stated intentions, I just can't wrap my head around the naivety. OpenAI corporate structure was a recipe for bad corporate governance. "We are the good guys here, the structure is needed to make others align with us."- an organization where ethical people can rule as benevolent dictators is the same mistake committed socialists made when they had power.
If it was that easy, AI alignment would be solved by creating ethical AI committed to alignment and giving it as much power as possible.
Altruists are normal humans. Nothing changes priorities faster than large sums of money. Any mix of ideals and profit-making must be arranged in a way that concerns don't mix. People in charge of ethics making life-changing money can't be expected to make those hard ethics over money decisions.
Bad organizational structure puts well meaning humans like Sutskever repeatedly into position where he must choose between wast sums of money or his ethical commitments.