Can AI Make Successful Video Ads?

 

As I prepared for the semester, I saw a mockup AI ad video created by a fellow classmate, Maggie Smoczynski. Her finished product blew me away! It led me to want to investigate the success of AI-generated video ads. Are they common practice now? How attainable are AI-generated videos for smaller companies? Are these campaigns at all successful?

Through my research on AI video ads, I came across an article written by AIContentfy titled AI-generated video ads: revolutionizing advertising. Below is a summary of the article and my thoughts on the subject.

The AIContentfy team explains that AI-generated video ads use AI technologies "to create and customize video ads for different audiences." AI-generated videos can analyze individual user preferences by using machine learning algorithms to get higher click-through and conversion rates. AI-generated video ads use several ML techniques, including natural language processing (NLP) and computer vision.

But why create a video ad through AI rather than the traditional channels? According to the AIContentfy team, there are four main reasons to consider.

Personalized Content. By using AI-generated video ads, marketers can personalize the advertising experience for each viewer. ML models can analyze large amounts of data regarding a viewer's interests, browsing history, demographics, behavior, etc. Imagine a dog food company targeting their ad to me by showing a video of a Bernadoodle (my dog), while a targeted ad for my sister might show a video of a Shih Tzu (my sister's dog). Conversion rates would surely increase for my family!

Efficient and Cost-Effective Ads. Traditional video creation can become costly: video equipment, actors, sets, costume changes, green screen, editing services, etc. The charges add up! The team at AIContentfy argues that AI-generated video ads are created "quickly and automatically, with minimal human intervention." However, an AI-generated video ad will only be as good as the data it's fed, so having access to a good database will be paramount for creating effective advertisements.

Precise Targeting. Thanks to the capabilities of these technologies, AI-generated video ads will identify very specific key audiences that are more likely to engage with the brand or product in question. 

Valuable Insights. Going hand in hand with the previous point, AI-generated video ads will also have access ti a much broader and deeper data set regarding its audience, including what types of advertisements work and which do not. 

Note: The AIContentfy article included examples of successful AI-generated video ads by Coca-Cola and Toyota. However, I was not able to find a video example of either after hours of searching the web.

Like everything regarding the fast-growing use and implementation of AI and ML tools in our everyday world, one must also consider the ethical implications of using AI-generated video ads. The team at AIContentfy highlights four primary areas of concern.

Privacy Concerns. To fulfill the promise of targeting potential customers based on their data, the AI must first collect this data from each viewer. The ethical implications surrounding what, how, when, and why of this data raise concern, especially if the viewers "feel that their data is being used inappropriately or without their consent." The Team and AIContentfy, as well as research I've concluded on my own, call for transparency to avoid this problem from arising.

Bias and Discrimination. My stepdad often says a program/app is only ever as good as the developers' abilities. Likewise, an ML algorithm can only be as good as its database. Skewed data will yield erroneous results. 

Responsibility and Accountability. The issues surrounding responsibility and accountability, according to AIContentfy, revolve around truthful and non-deceptive advertising. For a video ad created by AI, where does the responsibility lie to ensure an accurate depiction of the services or products advertised? Moreover, where does the accountability lie to ensure the advertisement follows the applicable laws and regulations of the market?

Manipulation and Transparency. Transparency seems to be the biggest potential issue, or most obvious solution, when it comes to ML algorithms and its uses. 

Again, their findings resonate with my research for a Business Analytics class I conducted with a classmate. Although our presentation leaned on the impacts on academics, the ethical implications apply to all instances where AI and ML intertwine in our everyday lives.