There is no such thing as perfect alignment. The notion that there is leads to misunderstandings like this: "By injecting inherent bias through the nation’s p

There is no such thing as perfect alignment. The notion that there is leads to misunderstandings like this: "By injecting inherent bias through the nation’s primary AI risk framework" All models are biased, the faster we appreciate this, the closer we can (imperfectly) align them to truth.