My reaction on Twitter: https://twitter.com/stuartbuck1/status/1719152472558555481
Stuart Buck
This post seems even more relevant and true now, in 2023.
This makes as much as sense (probably more) than anything Eliezer has said recently...
Here’s an interesting example of a fairly unique governance arrangement: https://issues.org/revisiting-nsf-nsb-science-advice-olds-rosenberg-robichaud/
Ostrom is great, obviously. In fact, I forgot how thoroughly I summarized a good bit of that literature in a 2001 piece: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=268744
I recently stumbled across a sociology classic of sorts: Paul J. DiMaggio and Walter W. Powell, “The Iron Cage Revisited: Institutional Isomorphism and Collective Rationality in Organizational Fields” (American Sociological Review, 1983). https://www.uio.no/studier/emner/matnat/ifi/INF9200/v10/readings/papers/DeMaggio.pdf
“We ask...why there is such startling homogeneity of organizational forms and practices.”
The main answer: “We identify three mechanisms through which institutional isomorphic change occurs...: 1) coercive isomorphism that stems from political influence and the problem of legitimacy; 2) mimetic isomorphism resulting from standard responses to uncertainty; and 3) normative isomorphism, associated with professionalization.”
I said on Twitter a while back that much of the discussion about “alignment” seems vacuous. After all, alignment to what?
The designer’s intent? Often that is precisely the problem with software. It does exactly and literally what the designer programmed, however shortsighted. Plus, some designers may themselves be malevolent.
Aligned with human values? One of the most universal human values is tribalism, including willingness to oppress or kill the outgroup.
Aligned with “doing good things”? Whose definition of “good”?