# What is GAN? Definition & Technical Mechanics

> **Source:** [SPIDITS AI AI Glossary](https://spidits.com/ai-glossary)  
> **Category:** Generative AI  
> **Canonical Citation:** [https://spidits.com/ai-glossary/gan](https://spidits.com/ai-glossary/gan)

## Definition
A Generative Adversarial Network (GAN) is a generative AI architecture consisting of two neural networks: a Generator (which creates fake data) and a Discriminator (which evaluates if the data is real or fake). The networks train in competition, forcing the generator to produce high-fidelity data.

## Architectural Mechanics & Deep Dive
A Generative Adversarial Network (GAN) is a generative model framework consisting of two neural networks: a generator (which creates synthetic data) and a discriminator (which evaluates authenticity). Trained together in a zero-sum game, the generator learns to produce highly realistic data to fool the discriminator, widely used for image generation and style transfer.

## Industry Applications & Real-World Use Cases
Image synthesis, deepfakes, texture generation, and style transfer

## Frequently Asked Questions

### What is mode collapse in GANs?
Mode collapse is a training failure where the generator learns to produce only a limited variety of outputs that fool the discriminator, rather than diverse samples.

### Who invented GANs?
Ian Goodfellow and his colleagues in 2014.

## Related Terminology
- [generative-ai](https://spidits.com/ai-glossary/generative-ai)
- [neural-network](https://spidits.com/ai-glossary/neural-network)

---
*From the SPIDITS AI Knowledge Base (336+ terms indexed). Explore full technical definitions: [https://spidits.com/ai-glossary](https://spidits.com/ai-glossary)*