4 ms·
> Instead they believe model alignment, trying to understand when a user is doing a dangerous task, etc. will be enough. In other words: motivated reasoning.
by ngcazz 1y ago
> Instead they believe model alignment, trying to understand when a user is doing a dangerous task, etc. will be enough.
In other words: motivated reasoning.