Google-Funded Company Fights Fake News Using AI

Source: eyerys.com

The battle over distinguishing what is real news and what is fake news has become a daily struggle.  People are bombarded with news on the television, radio, internet and through social media but how does one navigate through the falsehoods to arrive at what is true?  Apparently, there is a company that is funded by Google who is utilizing artificial intelligence to battle against fake news.

Fake News has Existed for Centuries

While there are people who feel that life was simpler centuries before the radio, telephones and other forms of technology like it that were ever invented, this idea partly stems from a time where fake news was not part of the main-stream vocabulary and people basically trusted the news that they read.  However, individuals recognized centuries ago that there was a problem determining what was real versus what was fake. 

Source: searchengineland.com

Writer Bernard Marr from Forbes recently posted an article about how this Google-funded company was trying to tackle the problem of fake news and pointed out how this issue existed centuries ago by citing something Jonathan Swift said over two centuries ago; Falsehood flies, and the truth comes limping after it.  Obviously, this problem has grown more excessive being we live in an age of how Twitter and Instagram alone can spread falsehoods that can reach people on a global scale.

Combating the Dilemma of Fake News

There is no denying that fake news has become a problem that occurs daily on both sides of politics.  This issue has become so big that Gartner has made the prediction that society will absorb more falsehoods than truth by 2022.  The dilemma is how to combat fake news and though technology has compounded the situation, there is new hope that it might offer the solution; particularly, in artificial intelligence.

Utilizing AI currently, in its most useful form, is already helping institutions in a variety of ways.  Machine learning is being used in banks and financial institutions to review financial transaction records for any noticeable signs of fraud or errors.  The information can then be used to make institutions more efficient without the need of human input.

Source: techcrunch.com

These same algorithms can be programmed to monitor media from news organizations and social networks; then, seek out any tell-tale signs in a story or comment that would possibly be out of sync with the objective truths already known about the events or situations.  Currently, an interesting application of this technology has already been developed. 

Belgium-based startup VeriFlix has developed an application that utilizes this type of technology.  Apparently, a method has been created where user-submitted videos are scanned and attempt to ascertain whether the they are real or fake; scanning videos submitted by users play an increasingly vital part in the output of a large portion of media organizations. 

The company’s technology acquired funding through Google’s Digital News Initiative and is currently being used by Rourlarta, which is one of the biggest media outlets of the country, which has shown promising results.  Founder Donald Staar spoke with Marr about how the evolution of the platform from the original conception of a peer-to-peer crowd-sourcing app utilized for videos.  Staar told him that once the videos get sent to the platform we add a layer which first detects the content of every stream – so we can say what we see in the video, alongside the geolocation data and time stamp.

He went on to say that and once the videos are tagged we can compare them to one another, so that if for example, one request results in 1,000 videos, we can compare the content of every video and if a majority of the videos show the same content, then it can verify the authenticity of what has been shot.   If 800 videos out of 1,000 show the same thing then the probability that the video has been faked is very low.

VeriFlix utilizes real-time object detection algorithms known as You Only Look Once (YOLO) to label and classify video contents before sending that information through to proprietary algorithms; these were created in a partnership formed with KU Leuven University.  The algorithms study the information along with geolocation and timestamp data which passes through the secure interface of the application.

Source: facebook.com

According to Staar, there are two main advantages – the first is that media companies can now make sure that videos they use are authentic and shot in the location where they say they are taken, and not modified or doctored.  The other advantage is that they are able to bridge the gap between themselves and their audience – let their audience become a part of the story, and source exclusive and verified content very quickly. It can be for small things, too – it doesn’t have to just be big, breaking news.

The battle to combat fake news will not be won overnight.  However, the technology that some say is at the heart of spreading fake news has the potential to stop it in its tracks for good.