Understanding Image Models and AI

What are image models commonly used for?

• Only for video games

• Image recognition and classification

• Language translation

• Spreadsheet calculations

What type of data do image models primarily work with?

• Text

• Numerical data

• Image data (pixels)

• Sound

How can you tell an image was created by AI?

• It's impossible to tell.

• The image will be very dark and gloomy.

• There will be a stamp on it that says it was made by AI.

• It is difficult, but look for parts of the image that are either too perfect or seem unnatural.

Generative AI models use a GAN. What are the two parts of a GAN?

• A Generator and a Neutralize

• A Graph and a Naysayer

• A Generator and a Discriminator

• A Creator and a Critique

How is an AI-created image different from a deepfake image?

• Both are created from AI, but a deepfake image is intended to fool people into believing something happened that didn't actually happen.

• They are both AI-created, so there is no difference.

• Deepfakes are used only for good purposes.

• They are not different; they both have the same intent.

The generator in a GAN is trained to fool the discriminator into believing that the images it generates are real.

• True

• False

What are the two components of a Generative Adversarial Network (GAN)?

• Creator and Evaluator.

• Generator and Discriminator.

• Encoder and Decoder.

• Simulator and Analyzer.

Can you tell which of the following images was created with AI?

• Left

• Right