Creative Content: Adversarial Patterns-AI-ML

A creative interpretation about adversarial patterns by digital artist Arzu Celik - Azul Creative Mds

Fool the system: Chatgpt vs Bard

Have you ever heard a scenario where the pattern recognition system faces an intelligent and adaptive adversary that modifies patterns to fool the system.

If you haven’t heard yet, In this article, we are going to focus on a general heading of a study called “Adversarial machine learning” with an interesting experience.

What is this interesting experience? In this experience, we are going to do some research about “Adversarial machine learning” by using 2 kinds of AI chatbot called ChatGPT and Bard- Google. However, before the test, we will find out the accurate contents by using Wikipedia and real articles (google searches) in order to give exact information.

To briefly describe: Adversarial examples are specialised inputs created with the purpose of confusing a neural network, resulting in the misclassification of a given input. These notorious inputs are indistinguishable to the human eye, but cause the network to fail to identify the contents of the image. There are several types of such attacks.

After explaining the content from some sources. Now, it is time to expand my researches by asking to ChatGPT and Bard.

Firstly, I asked this question to ChatGPT:

Reach the full article via Medium Link:

https://medium.com/@arzucelik-acm/fool-the-system-chatgpt-vs-bard-ead9f837d006?sk=dc653aeb21d0a80884c318e376212ca0

You may also like