Following the news about Facebook working on a censorship tool for China, Microsoft has confirmed its Chinese-language chatbot filters certain topics.
Microsoft’s Chinese chatbot, Xiaoice, has over 40 million users. Reports last week suggested that the bot would not directly respond to questions that might be deemed sensitive by the Chinese state. Introduced in early 2014, Xiaoice uses artificial intelligence - mining Chinese websites to gather material for conversations - to respond to questions about almost every topic, except for those the state might not approve of. Tiananmen Square massacre of 1989 and similar other topics would force the chatbot to offer evasive responses, reports had claimed. Along with China's President Xi Jinping, overthrowing the communist parties, the chatbot also interestingly avoids answering questions about the US President-elect Donald Trump.
Xiaoice replies with "Am I stupid? Once I answer you'd take a screengrab" or similar answers to these questions.
Microsoft confirms filtering some topics to create "the best experience"
Microsoft has now officially confirmed that Xiaoice is filtering certain topics. "We are committed to creating the best experience for everyone chatting with Xiaoice," a Microsoft spokesperson told Fortune. "With this in mind, we have implemented filtering on a range of topics." The software maker didn't expand on the topics this filtering applies on.
China routinely blocks websites like Google and Twitter, along with heavily censoring content that is considered politically sensitive by the state. Facebook, the largest social networking site, remains blocked in the country since 2009. Reports last week revealed that networking giant was building a censorship tool to gain a re-entry in the country. The software is designed to suppress posts from appearing in people's news feeds in specific geographic areas.
While it's unknown if Facebook plans to hand over the tool to China, the country keeps getting the worst score in online freedom. China also recently passed a new cybersecurity law that will go into effect in June, next year. Economists and security analysts are worried that the new law will only create more barriers for international businesses in China, and cripple online freedom even further.
