AI is one of the most hotly debated topics at the moment. Month by month we are seeing AI technology advance at a startling pace. And, like with any new technology, there are a lot of discussions about the potential dangers of AI. Many nations are still very behind on passing proper AI laws and regulations. Meaning, for the most part, AI development and usage is currently unimpeded.

Today we want to focus on the latest viral trend hitting TikTok and other social media platforms. AI generated videos. You will have doubtless seen hundreds of these if you have spent any amount of time scrolling through TikTok. But just how harmful are these AI videos?

AI Unaware

You might be thinking to yourself that you haven’t actually seen that much AI generated content circulating on social media yet. But, especially on TikTok, chances are you weren’t even aware that the video you just watched was made by AI.

When people think of AI generated content they think of the surreal images made by DALL-E or the creepy TikTok time traveler videos where everything looks like it is bending and melting. These are the more overt types of AI content being produced.

But often you might be watching a TikTok video where the app’s own text-to-speech function is narrating a story written by the poster. Or relaying information over a background of lovely nature scenes. Here’s the bad news. A considerable percentage of these types of videos are AI generated. It is all too easy for influencers to outsource their content creation to AI programs and their audience be none the wiser.

Stolen Content

You might be thinking that the idea of AI generated fictional content isn’t all that harmful. And you would be right to a certain extent. The fact that we can create such wonderful works of art with AI is a testament to just how far humanity has come. But the issue comes from how AI learns to make this art.

AI programs need a lot of data to be able to create something. For example, if you wanted an AI to make you a picture of a dog you would need to show it hundreds of pictures of a dog first. So where do modern AI programmes learn their artistic skills? Sadly it is from artists whose work is publicly available on the internet.

The effect of this is twofold. Firstly, it means AI is essentially allowing some content creators to plagiarize other people’s work without crediting them. Secondly, it runs the risk of causing the creative industries to collapse. Why pay for a commissioned art piece if you can have an AI program make one for you in the style of your favorite artist?

A Force For Good

But it isn’t all bad news for honest content creators. A lot of the most popular content creators are now using AI to enhance their original content. I’ve been using a TikTok follower counter to track the uptake in followers smaller pages have been gaining since they started using AI to help make their art. And I have noticed a sharp increase in followers and follower engagement.

AI is also being used by TikTok and other social media platforms to better moderate what content is being posted. Previously, social media sites needed people to sift through all the illegal content people would try to post. And it is no understatement that this line of work can be very mentally scarring. Now we can use AI to deal with the more technical aspects of internet moderation.

Deep Fakes

But even if it can be used for good, there is still a lot of harm that can be caused by AI content. You may have heard about deep fakes. It is the process of digitally altering an image or video to make one person look like another. Often for comedic purposes. But, more and more, it is being used as a means for spreading misinformation and deceit.

AI has become very good at creating convincing vocal imitations of almost anyone. I have seen hundreds of comedic videos of Donald Trump and Joe Biden playing video games together. But imagine if, instead, we were shown videos of a respected politician making racist remarks? Or a celebrity admitting to a crime they never committed? This is the primary reason that more countries need to bring in stricter AI legislation. But some argue that it is already too late. Now that the technology is publicly available is it possible to stem the damage that could be caused?