skip to content
Usarral
Table of Contents

Have you ever wondered how fake news and machine-generated texts affect the Artificial Intelligences (AIs) behind the technology we experience today? Recently, I had an interesting conversation with a coworker about this topic, and it honestly left me reflecting.

The Conversation that Started It All

It all began with a question my coworker posed: “How can automatically generated articles by AI and fake news affect the training of new AIs?” He was concerned that these digital intelligences might be spreading false information like confetti. And yes, I agree with that concern.

Ethical Challenges in the World of AIs

Then we got to the crux of the matter: the quality of data.

Just as our opinions are influenced by what we read, AIs are influenced by the quality of the information they receive. It’s like giving them a recipe for a cake and expecting it to turn out perfect. But of course, if the recipe is wrong, the cake will be too.

The conversation led us to the importance of correcting AIs when they make mistakes. Feedback is key, but we also mentioned something as fundamental as transparency. How can we trust responses if we don’t know where they come from?

The Ethical Dilemma of Censorship in Data Training

As our conversation deepened, we delved into an ethical dilemma surrounding censorship in the data training process. Filtering information during this initial phase might sound like a solution to ensure quality, but here comes the big question: Where do we draw the line between improving quality and censoring valuable perspectives? If we add censorship, the AI won’t get real information but rather the information we want it to give.

Actions We Can Take

The conclusion of our conversation was that we all have a role to play. From using high-quality datasets to patiently correcting AI errors, to being honest about when a text is machine-generated. And, of course, educating ourselves! Knowing how to identify false information is crucial. Additionally, it’s increasingly important to cross-check information and not settle for a single perspective, as an AI won’t always be right.

My Personal Point of View

So, to summarize my opinion, I believe this world of AI needs an ethical touch. If we are going to let machines learn from us, we must ensure they learn the best from us. Let’s use high-quality data, correct their mistakes patiently, be honest in our interactions with them, and promote values like truth, justice, and fairness in AI development. Only then can we create a digital future that is prosperous and fair for all.