• @u_tamtam
    link
    17 hours ago

    Isn’t that the essence of the issue, that those models are loaded with biases, that might or might not overlap with dominant ones in inscrutable ways, hence producing new levels of confusion and indirection?