I make a contribution to AI safety that is endorsed by at least one high profile AI alignment researcher by the end of 2026