Artificial intelligence models developed by Microsoft and Alibaba have, for the first time, outperformed humans in a reading comprehension challenge.\nThe Stanford Question Answering Dataset (SQuAD) consists of a series of questions to which the answers can be found within more than 500 Wikipedia entries.\nAlibaba\u2019s deep neural network model scored 82.440 on the \u2018exact match\u2019 part of the test, besting the scores achieved by humans (82.304). Microsoft\u2019s similar model achieved a score of 82.650.\nThe scoreboard is a who\u2019s who of corporates carrying out artificial intelligence research, featuring the likes of Google, IBM Research, Facebook AI Research, Salesforce Research, Tencent and Samsung.\nAlibaba and Microsoft have been placed joint first in the ranking, although both companies claim to have reached the better-than-human milestone first.\nWhile Microsoft is listed as having registered its score on January 3 and Alibaba two days later, Alibaba said those dates were when the companies submitted their models, not when test results were registered.\n\u201cIt is our great honour to witness the milestone where machines surpass humans in reading comprehension,\u201d said Luo Si, chief scientist for natural language processing at Alibaba\u2019s Institute of Data Science and Technologies (iDST) in a statement. \u201cWe are thrilled to see NLP research has achieved significant progress over the year. We look forward to sharing our model-building methodology with the wider community and exporting the technology to our clients in the near future.\u201d\nMing Zhou, assistant managing director of Microsoft Research Asia, said despite the milestone, overall, people are still much better than machines at comprehending the complexity and nuance of language.\n\u201cNatural language processing is still an area with lots of challenges that we all need to keep investing in and pushing forward,\u201d he said. \u201cThis milestone is just a start.\u201d\nThe big AI players are investing heavily in reading comprehension and response models.\nAlibaba said it had been using the underlying technology during its \u2018Global Shopping Festival\u2019 for a number of years to answer customer inquiries.\nMicrosoft said it was applying earlier versions of the model to its Bing search engine.\n \n\u201cThese tools also could let doctors, lawyers and other experts more quickly get through the drudgery of things like reading through large documents for specific medical findings or rarified legal precedent. The technology would augment their work and leave them with more time to apply the knowledge to focus on treating patients or formulating legal opinions,\u201d the company wrote in a blogpost.\nIt is also working on models that answer probable follow-up questions.\n\u201cFor example, let\u2019s say you asked a system, \u2018What year was the prime minister of Germany born?\u2019 You might want it to also understand you were still talking about the same thing when you asked the follow-up question, \u2018What city was she born in?\u2019\n\u201cIt\u2019s also looking at ways that computers can generate natural answers when that requires information from several sentences. For example, if the computer is asked, \u2018Is John Smith a U.S. citizen?\u2019 that information may be based on a paragraph such as, \u2018John Smith was born in Hawaii. That state is in the US\u2019\u201d Microsoft explained.