Davidmanheim answers Resources on quantifiably forecasting future progress or reviewing past progress in AI safety?