I’ve been very impressed with MIRI’s output this year, to the extent I am able to be a judge. I don’t have the domain specific ability to evaluate the papers, but there is a sustained frequency of material being produced. I’ve also read much of the thinking around VAT, related open problems, definition of concepts like foreseen difficulties… the language and framework for carving up the AI safety problem has really moved forward.
Thanks! Our languages and frameworks definitely have been improving greatly over the last year or so, and I’m excited to see where we go now that we’ve pulled a sizable team together.
Donation sent.
I’ve been very impressed with MIRI’s output this year, to the extent I am able to be a judge. I don’t have the domain specific ability to evaluate the papers, but there is a sustained frequency of material being produced. I’ve also read much of the thinking around VAT, related open problems, definition of concepts like foreseen difficulties… the language and framework for carving up the AI safety problem has really moved forward.
Thanks! Our languages and frameworks definitely have been improving greatly over the last year or so, and I’m excited to see where we go now that we’ve pulled a sizable team together.