There is a growing concern about fake news after the elections with experts saying that Donald Trump's election was due, in part, to these fake news. Hillary Clinton is now calling an action against them saying they have 'real world consequences.' The tech world has heeded that and now a group of hackers, researchers, engineers, and academics have converged to tackle the issue with the help of artificial intelligence.
Despite the advance algorithms they have been using, Google and Facebook failed to detect fake news. According to Quartz, an AI that's considered state-of-the-art at the moment, can only respond to question and answer games. However, this is not enough if you want to need to check the factuality of an information.
According to Richard Socher, chief of Salesforce Research, said that with the novelty of news going around the Internet, even the most powerful AI will have difficulty reading and understanding all the information that's being published.
When it comes to fact-checking, us humans are always the best. However, even humans don't agree on alll issues tackeled at all times.
There have been several AI projects that are trying to combat the scourge of fake news. One was a tool called FiB, which won a Princeton University award. The software reads and analyzes news posted on Facebook against reputable websites. If the post contains images or tweets, the software converts it to text.
Another proposal comes from Ozlo, a startup which developed a personal assistant that analyzes restaurant reviews. What the company did was teach its algorithm what to trust and what not to trust. The company hopes to expand beyond the restaurant industry using its AI.
Despite these efforts, they can't totally eliminate fake news, only limit it. There's still a lot to be done in order to develop an AI that could perfectly and efficiently spot fake news from not. For now, it still boils down to the readers to spot which is true and which is not.
