Damilola was surfing through the internet with her mom when, suddenly, her mom jerked up. She had just seen a video of a crocodile apparently eating a woman who was feeding it.
Her mom could swear that she had just watched a woman almost get eaten alive by a crocodile in real time. This sparked an argument between the both of them. Damilola told her that the video was clearly AI-generated, partly because the video itself looked suspicious and the caption was badly written. Her mom still had her doubts about whether the video could actually be real, but eventually, she was convinced that it wasn’t.
So, what is the point of telling this story?
A lot of people have been conditioned to believe that what they see online is reality, when, in fact, it may not be. We are gradually becoming accustomed to assuming that the majority of the videos we come across are genuine, when some of them are not.
The scary part is that AI-generated content is becoming increasingly convincing. Not everyone can immediately tell the difference between something created by artificial intelligence and something that actually happened. This is particularly concerning in a world where people consume information primarily through social media and other digital platforms.
This is where regulation comes in.
There should be policies that encourage transparency in the use of artificial intelligence, particularly where AI-generated or manipulated content could reasonably be mistaken for reality. One possible approach is requiring clear disclosure or labelling of AI-generated content. If something has been created or substantially altered using AI, people should not have to guess.
But the problem goes beyond misleading videos.
We have also seen people use AI to design flyers, posters and other forms of communication, sometimes with results that completely miss the intended message.
Take, for instance, a flyer Damilola came across online. At first glance, she almost thought it was an obituary flyer. She had to read closely to convince herself that it wasn’t an RIP post. Lo and behold, it was actually a birthday flyer.
Damilola was honestly upset because how does a birthday flyer look like an obituary?
This brings us to another issue: the importance of prompts.
AI can only work with the instructions we give it. If you are not going to be detailed about how you want your flyer, image or design to look, then please don’t leave everything to your AI tool’s discretion. You might ask for a birthday flyer and end up with something that looks like a memorial service announcement.
The issue, therefore, is not necessarily that AI should not be used. AI is an incredibly useful tool, and there is no doubt that it has changed the way we create, communicate and work. The problem arises when we allow AI-generated content to blur the line between reality and fiction or communicate something entirely different from what was intended.
More concerningly, AI-generated content can be used to create false narratives, manipulate people’s perceptions and, in some cases, damage relationships. Technology should make communication easier, not make us question everything we see and hear.
Ultimately, AI should be used in a way that complements reality rather than creates a flawed expectation of it. As artificial intelligence continues to develop, policy enforcement—not merely policy enactment—will be essential in ensuring that AI is used responsibly.
The question is no longer simply, “What can AI do?”
Perhaps we should also be asking, “What should AI be allowed to do, and how do we ensure that people know when they are looking at something created by AI?”
At the Technology Law Club, Redeemer’s University, we discuss pressing issues arising from the use of technology and its impact on society, law and everyday life.
If you would like to be part of this community and join conversations like this, send us an email at runtechlawclub@gmail.com