Thank you for your comment! :)

I agree with you that we should not only try to achieve AGI, but constantly assess the ethical aspects of this tech. However, I'm not sure how could we define an ethics test for these AIs. (If you have any info I'd love to know!)

Cheers :)