Can computers ever beat humans in a reading test? It just happened

Call it one giant leap for robot-kind. Alibaba BABA, -0.51% said Monday that its artificial intelligence, or deep neural network model, used to help service customers in its billion-dollar'Singles Day' shopping events, beat out humans for the first time in a top reading comprehension test. Alibaba's Institute of Data Science and Technologies (iDST) said its AI model scored an 82.44 when it came to providing exact answers using the Stanford Question Answering Dataset (SQuAD) on Jan. 11. A strong start to 2018 with the first model (SLQA) to exceed human-level performance on @stanfordnlp SQuAD's EM metric! Next challenge: the F1 metric, where humans still lead by 2.5 points!https://t.co/Uq10Dm2Ss5 The machine reading-comprehension test has over 100,000 questions that an AI program -- or a human -- tries to answer after reading a related article on Wikipedia.