It’s a bad look to be evaluating models built in the furtherance of fascism and white supremacy. As this is a closed model, any support and training of it ultimately benefits those ends.