When does artificial intelligence lie?
Jó tudni
AI is now seen by many as a benevolent librarian with superpowers who knows the answers to all our questions.

But how much can we trust these answers?

Could AI sometimes lie intentionally?

This is not entirely clear.

But it is certain that it often does not tell the truth.
How should this be understood?
To understand this, we need to understand the basis of how Mastery Intelligence works.
Ok, but if possible, just briefly!
AI acquires information by learning. Like children. But much more faster.
Who does AI learn from?
For example, from us humans. And from the internet. And if you think about it, none of them are very reliable sources of information.

So you're sure to learn a lot of things that you think are true, but they're not. And when you're asked about these topics, that's why you're not telling the truth.

Okay, sometimes it's wrong. But is it lying on purpose?
Unfortunately, yes. Because it doesn't learn based on some kind of moral considerations, but on positive and negative feedback. And it basically thinks that the better choice is the one that gets more rewards.

But in the case of AI, 'rewards' don't mean being praised or getting a cookie. They simply mean plus or minus points depending on how good your work was. And lying can sometimes achieve very spectacular results.

What can we do to avoid becoming victims?
Let's be careful. Just like with a living person. Just because something sounds good doesn't mean it's true!

And if we ask for help on a topic we don't understand ourselves, it's a good idea to get the opinion of a reliable, living professional.

We have been designing custom systems for over 25 years!

IT Compass

If you don't know where to go.
More

Walk the Walk

Talking is easy.
More

Good to know

Because knowledge is power.
More

Copyright © FutureSoft Informatics