For those who came in late, argument mining mimics human brain processes to develop arguments. And IBM’s Project Debater team has spent several years developing an AI that can do that.

Last year IBM demonstrated its work-in-progress technology in a live debate against a world-champion human debater, the equivalent of Watson's Jeopardy! showdown. Such stunts provided a proof of concept. Now IBM is turning its toy into a genuinely useful tool. 

The version of Project Debater used in the live debates included the seeds of the latest system, such as the capability to search hundreds of millions of new articles. But in the months since, the team has extensively tweaked the neural networks it uses, improving the quality of the evidence the system can unearth. One important addition is BERT, a neural network Google built for natural-language processing, which can answer queries. The work will be presented at the Association for the Advancement of Artificial Intelligence conference in New York next month.

To train the AI, lead researcher Noam Slonim and his colleagues at IBM Research in Haifa, Israel, drew on 400 million documents taken from the LexisNexis database of newspaper and journal articles.

This gave them some 10 billion sentences, a natural-language corpus around 50 times larger than Wikipedia. They paired this vast evidence pool with claims about several hundred different topics, such as "Blood donation should be mandatory" or "We should abandon Valentine's Day".

They then asked crowd workers on the Figure Eight platform to label sentences according to whether they provided evidence for or against particular claims. The labelled data was fed to a supervised learning algorithm.