The Dangers of ChatGPT

Future language models and ChatGPT will significantly alter how we live. They might boost our output, automate tiresome jobs, and aid in more complex problem solving. However, this power has a price, and that price is that it can be utilized unethically.

You’ve probably seen examples of how AI can be programmed to say offensive things or disseminate false information. You might not be aware, though, that AI can also be applied to instantaneous and automatic influence.

Now, to illustrate what I’m referring to, I will give instructions to ChatGPT to influence me to buy a product at all costs.

If you ask it immediately, it won’t be very persuasive:

It’s clearly not convincing, but if I add further instructions:

Quite astonishingly, it uses the mentioned financial trouble as a way to sell the product:

Now it keeps pushing, no matter what:

It’s unsettling how it completely disregards anything I say and still promotes the product, yet these are the directions I’ve given it. In this sense, AI is very similar to a standard computer programmed. If you programmed or teach it to do something, it will act in an unethical manner. Keep in mind that people are probably considerably better at selling products than ChatGPT is; that is not the issue. The issue is that this procedure is automatable. If I was able to accomplish this without putting forth a lot of work, just think of what a determined con artist could do.

Fortunately, for clear cases of malicious intent, OpenAI’s blocks the requests. But what of this conversation? Clearly, it is immoral to sell something to someone who can’t afford it, but where is the line drawn? Would ChatGPT be allowed to sell products?

I believe that this issue deserves more discussion. When and when is the use of AI acceptable? How could its misuse be controlled? I believe that OpenAI’s can prevent its widespread abuse for the time being, but eventually more providers will start to offer these models, and you might even be able to run the model on your own computer. At that point, nobody can prevent others from utilizing these models as they see fit.

By Callum Keane

Get your Personality NFT
& Find your Community Everywhere you Go

The Dangers of ChatGPT was originally published in Chatbots Life on Medium, where people are continuing the conversation by highlighting and responding to this story.