Published On: Sat, May 12th, 2018

Artificial Intelligence and the Phenomenon of Fake News: Detection versus Creation

Last month, Mark Zuckerberg, the chief executive Facebook, was questioned for over ten hours by Congress. While the government was interested in finding out how Facebook´s privacy practices protected their clients from possible data theft and confidentiality issues, the issue of “fake news” spreading across the internet also came up.

photo/ Gerd Altmann via pixabay

Zuckerberg mentioned that artificial intelligence technology offers “a scalable way to identify and root out most of this harmful content” that characterizes some of the billions of Facebook posts uploaded every day. Is he right that artificial intelligence might be able to act as a filter for the fake news stories that make their way onto our computer and cellphone screens, or could artificial intelligence actually be used to increase the amount of fake news content floating around the web?

WorkFusion, a leader in the world of intelligent automation, says that cognitive automation “is software with the ability to perform more complex work that involves unstructured data (like images, documents, or PDFs).” Cognitive Automation is powered by machine learning, and a basic example of this type of AI technology is seen in almost everyone´s email account.

Machine learning algorithms have been employed for years to help fight the phenomenon of spam email. Instead of having 200 messages in your inbox every time you check your email, these algorithms allow AI technology to analyze the text of a given message and establish a likelihood that the message is a real communication that will be useful to you instead of a piece of junk mail trying to get you to purchase something.

Fortunately, these AI algorithms that work to help keep your email account organized (and more secure) are also useful in the fight against fake news. What Mark Zuckerberg was referring to is the ability for AI systems to assess the headlines or the text of a given news article and then compare that article to what other reputable news media has published. This could help Facebook, and several other websites, detect certain websites and social media accounts that are actively scattering fake news around the internet.

On the other hand, there are certainly dangers related to this technology, as AI technology might pick up on the political bias and social prejudice that tend to characterize so much information on the internet. For example, it is possible that AI software put in place to detect fake news sources could potentially limit the free expression of specific points of view that are not widely held. Fortunately, as AI technology advances, it should be able to address these issues.

Another problem with AI and the phenomenon of fake news is that many people are already using machine learning systems to manipulate the news. “Deepfakes” go far beyond traditional “Photoshop” and have the ability to create photos and videos that can be close to impossible to detect reality from make believe.

Researchers at the University of Washington recently created this video of former president Obama. While the video certainly looks and sounds exactly like how Obama talked, the team actually used a neural network AI to model the shape of Obama´s mouth.

While all of us have a responsibility in responsibly sourcing the news we use to form our opinions, as AI systems continue to develop, they might play a major role in helping to sift out the worse forms or fake news that contaminate our political systems and confuse the general public.

Author: Click Tech Tips

photo/ Gerd Altmann via pixabay

On the DISPATCH: Headlines  Local  Opinion

Subscribe to Weekly Newsletter

* indicates required

About the Author

- Outside contributors to the Dispatch are always welcome to offer their unique voices, contradictory opinions or presentation of information not included on the site.

Leave a comment

XHTML: You can use these html tags: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>



The Global Dispatch Facebook page- click here

Movie News Facebook page - click here

Television News Facebook page - click here

Weird News Facebook page - click here