Fake news are plagues on the internet. With no single solution to end them, companies are trying to deliver their own method. One of which, is RobHat Labs.
The goal is to spot the fakes before internet users share them.
Using machine learning algorithms and textual analysis, SurfSafe works by determining the trustworthiness of content based on online text and image data.
By understanding the history of the image and whether an image has been manipulated or used out of context, SurfSafe provides the indicators that inform users whether or not the online contents they see are fakes, with links for more information.
According to Ash Bhat, co-founder and CEO of RoBhat Labs:
"People are misled every day and reputable news outlets are having their trust tarnished because of the actions of a few. While our Botcheck extension discovers suspicious Twitter sites, our users asked us to develop a solution that would help them fight against fake news across every browsing experience. SurfSafe accomplishes this by warning users when online content is trying to hijack your beliefs."
"We want SurfSafe to become a solution that’s analogous to antivirus software. We want to scan your news feed for fake news as you browse."
Initially, SurfSafe is available as extensions on Google Chrome, Opera, and Firefox browsers.
The extension is created only to detect hoax images. The creators say this is an acceptable limitation, since its goal is to the stop the viral spread of fake news.
Since the extension uses machine learning algorithms, it should get smarter the more the user uses the plugin. Bhat said that the average internet user sees about hundreds of thousands of images a day. The plugin saves the signature of all of these images, looking for subtle variations to the fingerprint, or hash, that accompany even minor edits.
Along with providing the public with powerful tools that can be used to determine fake news, RoBhat Labs intends to publish insights and warn the public regularly about the misinformation being discovered on the internet.