Using AI to Identify Fake News

To combat the spread of misinformation and fraudulent news online, social media giants have sought the help of startups specializing in Artificial Intelligence to help them identify and flag fake news.

Using AI to Identify Fake News
Combating Fake News through AI

As fake news keeps spreading its tentacles amidst the ever-increasing online database, it’s become increasingly harder to separate them from fact based information. Fake news usually relies on sensationalized stories that draws a reader’s attention, making profit through a process known as clickbait. The term gained widespread acceptance in the past couple of years, fueled by the spread of fraudulent political stories aimed to scandalize Hillary Clinton, the democratic candidate for the 2016 American presidential election. Such stories were spread mainly through social media platforms, and ultimately resulted in a shift in people’s perception that helped elect Donald Trump as the president. Since then, several prominent social media giants and tech agencies have reviewed and researched various ways to minimize their spread. As it’s impossible for humans to shift through millions of news stories posted online every day, they have turned to Artificial Intelligence (AI) to do the job for them.

The rise of new deep learning and machine learning algorithms have resulted in unprecedented rise of AI. It has proven to surpass human intelligence by beating them in online games, chess and cognitive thinking capacity. It’s no wonder that tech companies and news agencies are turning their head towards AI to help them identify and flag fake news across the internet. AI can train itself based on web score pages, facts verified from multiple sources, reputation of publishing website and use of sensational words to identify propaganda based news. After browsing through millions of new posts that pop up on the internet every day, it doesn’t have to delete the information, just have to flag it as being fake to aid its readers view the credibility of the material they are reading online.

Majority of fake news is spread through social media sites online. Facebook has been at the center of the controversy, when it was discovered that more than 70% of the fake news in the 2016 American presidential campaign was spread through its users. To circumvent this problem, Facebook is all set to acquire the UK based AI startup Boomsbury for $30 million. Bloomsbury, which primarily focuses on developing machine learning technologies and Natural language Processing (NLP) methods, can develop machines that could aid Facebook flag its content based on knowledge-based functionality.

Other tech behemoths like Google and Microsoft are also joining in on the race to acquire AI startups that could help them combat fake news on their websites. Millions of internet users get their information from their search engines on a daily basis. It’s important that the information that surfaces should be real and based on factual information. Google recently acquired DeepMind, another UK based AI startup to integrate cutting edge machine learning algorithms into their product. Microsoft has, too, acquired a Californian based startup, Bonsai in order to assist fight fake news.

As tech giants keep acquiring AI based startups to aid their fight against fake information, other startups have also played their part in advancing the technology. launched the beta version of their AI based software in December 2017 to detect phony stories nudity, malware and other kind of content that adulterates online information. As it becomes harder to detect fake news, as they keep evolving to sound more like real news, it’s only on the hands of advanced AI software’s to increase the credibility of online information.