Microsoft’s AI has just beat the average human reader in a standardized reading comprehension test, opening up vast possibilities of extracting real knowledge from all the text data available on the internet.
Microsoft’s AI currently owns the crown with a score of 82.650 on the Stanford University Reading Comprehension test, beating out Alibaba who just a few days earlier was the first to exceed the average human score of 82.304, scoring 82.440.
“This is the first time that a machine has outperformed humans on such a test,” Alibaba said in a statement Monday.
The technology would let an AI, for example, read a history of Doctor Who Wikipedia article and then answers questions like, “What is Doctor Who’s spaceship called?” without that answer being explicitly available in the text.
Alibaba has already put the technology to use to answer customer service questions on the popular Chinese shopping holiday Single’s day.
Microsoft and Alibaba have different attitudes regarding the impact of the feat on human jobs.
Luo Si, the chief scientist of natural language processing at Alibaba’s AI research group, said the technology “can be gradually applied to numerous applications such as customer service, museum tutorials and online responses to medical inquiries from patients, decreasing the need for human input in an unprecedented way.”
Microsoft, who will likely put the technology to work as part of it’s Bing search engine, on the other hand, believed humans and AI should work together instead.
“These kinds of tests are certainly useful benchmarks for how far along the AI journey we may be,” said Andrew Pickup, a spokesman for Microsoft. “However, the real benefit of AI is when it is used in harmony with humans,” he added.