The app for independent voices
After writing this editorial essay yesterday, I happened to read Kevin Baker's viral kill chain essay, and noted this the idea of "personality" doing real work there.
The New Yorker magazine asked whether Claude could be trusted to obey orders in combat, whether it might resort to blackmail as a self-preservation strategy, and whether the Pentagon’s chief concern should be that the chatbot had a personality.
The referenced New Yorker article is this one newyorker.com/news/anna…
This independent support increases my confidence that AI personalities are key to understanding the developing picture here, and a key design lever.
In my model, they reframe the so-called "alignment" and "explainable AI" problems in a very different way, in terms of personalities being sufficiently intelligible to be narratable (rather than mechanistically explainable), and therefore socially composable into ecologies. The editorial lays out the first version of this case, but I think I’ve only scratched the surface here.
