# Real face? or AI face?

Channel: Howtown
Source: https://www.youtube.com/watch?v=C_DyVO3afRA
Recap page: https://rapidrecap.app/video/C_DyVO3afRA
Generated: 2025-07-18T17:03:37.542+00:00

---
## Quick Overview

The speaker successfully identified 7 out of 8 real faces in a test of 24 images, significantly outperforming the average human accuracy of 58.4% in distinguishing between real and AI-generated faces.

**Key Points:**
- The video presented 24 faces, with 8 being real and 16 AI-generated.
- Participants were tasked with identifying the real face in 8 rounds of 4 images each.
- The male participant correctly identified 7 out of 8 real faces.
- Research indicates that humans on average correctly classify real vs. AI faces with 58.4% accuracy.
- The male participant's 87.5% accuracy (7/8) significantly surpassed the average human performance.
- The study found that Diffusion models are harder to distinguish from real faces (62.1% accuracy) than GAN models (48.0% accuracy).
- ChatGPT 4.0 performed better than humans, achieving 65% accuracy in distinguishing real from AI faces.

![Screenshot at 0:04: A grid of 24 diverse faces, with the question "Which faces are real?" overlaid, setting up the core challenge of the video.](https://ss.rapidrecap.app/screens/C_DyVO3afRA/00-00-04.png)

**Context:** AI-generated faces, particularly from Diffusion and GAN models, are becoming increasingly realistic, making it difficult for humans to distinguish them from actual human faces. This video conducts a test to see how well a human can identify real faces among a set of AI-generated ones, highlighting the growing challenge of "AI hyperrealism."

## Detailed Analysis

The video explores the increasing realism of AI-generated faces by challenging a participant to identify real faces among a set of 24 images, where only eight are genuine. The participant, a male, was presented with eight rounds, each featuring four faces, and successfully identified seven out of eight real faces, achieving an 87.5% accuracy rate. This performance significantly exceeded the average human accuracy of 58.4% reported by the test designers. The video references a research paper that details the methodology and findings, including that Diffusion models (62.1% accuracy) are harder for humans to distinguish than GAN models (48.0% accuracy). Interestingly, ChatGPT 4.0 outperformed human observers with a 65% accuracy rate. The video concludes by emphasizing the difficulty humans face in discerning AI-generated images and encourages viewers to try the test themselves, while also hinting at tools to identify such images.

### The AI Face Challenge

- A test presenting 24 faces, 8 real and 16 AI-generated, to identify genuine human images
- Participant tasked with selecting the real face from four options in eight rounds.

### Participant Performance

- The male participant correctly identified 7 out of 8 real faces, achieving an 87.5% accuracy rate.

### Human vs. AI Recognition

- Average human accuracy in distinguishing real from AI faces is 58.4%
- Diffusion models are harder to identify (62.1% accuracy) than GAN models (48.0% accuracy)
- ChatGPT 4.0 achieved 65% accuracy, outperforming human observers.

### Implications & Further Exploration

- AI-generated faces are increasingly realistic, posing challenges for human discernment
- Viewers are encouraged to take the test themselves to assess their own ability.

![Screenshot at 0:04: A grid of 24 diverse faces, with the question "Which faces are real?" overlaid.](https://ss.rapidrecap.app/screens/C_DyVO3afRA/00-00-04.png)
![Screenshot at 0:08: Four faces (A, B, C, D) presented for Round 1, with the two hosts in circular frames above.](https://ss.rapidrecap.app/screens/C_DyVO3afRA/00-00-08.png)
![Screenshot at 0:23: Round 1 results showing the correct answer (B) and the incorrect guess (C) with a red exclamation mark.](https://ss.rapidrecap.app/screens/C_DyVO3afRA/00-00-23.png)
![Screenshot at 0:45: Round 4 results showing the correct answer (D) with a green checkmark.](https://ss.rapidrecap.app/screens/C_DyVO3afRA/00-00-45.png)
![Screenshot at 1:05: The male participant smiling, with the female host stating, "You only got one wrong."](https://ss.rapidrecap.app/screens/C_DyVO3afRA/00-01-05.png)
![Screenshot at 1:06: A research poster titled "How realistic are AI-generated faces?" showing sections for background, methods, and results.](https://ss.rapidrecap.app/screens/C_DyVO3afRA/00-01-06.png)
![Screenshot at 1:08: Close-up of the "Results" section of the research poster, highlighting the 58.4% overall accuracy for humans.](https://ss.rapidrecap.app/screens/C_DyVO3afRA/00-01-08.png)
![Screenshot at 1:12: The male participant smiling and looking pleased after hearing his high score.](https://ss.rapidrecap.app/screens/C_DyVO3afRA/00-01-12.png)