AI Faces Research

Can you tell AI-generated faces from real ones?

AI-generated faces are becoming increasingly realistic. In many cases, people struggle to distinguish them from real faces — often performing no better than chance.

AI vs real face comparison illustration
Person viewing multiple faces as part of recognition task

The task

Decide whether each face is real or AI-generated.

In this task, you will be shown a series of faces and asked to decide whether each one is a real photograph or generated by artificial intelligence.

The images are designed to be challenging. Many AI-generated faces now appear highly realistic, making it difficult to rely on obvious cues.

Your responses help researchers understand how people detect synthetic faces — and why some people perform better than others.

Why it matters

AI-generated faces are already part of everyday life.

Synthetic faces are now widely used in social media, marketing, and online identities. While some uses are harmless, others contribute to deception, fraud, and misinformation.

As the technology improves, it becomes increasingly difficult to tell real faces from artificial ones.

Understanding who can detect AI-generated faces — and who cannot — is important for improving digital security and resilience.

Read more in the UNSW Newsroom →

Individual differences

Some people are much better at detecting AI faces.

Higher ability

Super AI-detectors

Some individuals consistently outperform others when distinguishing real faces from AI-generated ones.

These individuals may rely on subtle visual cues or patterns that others miss.

Typical performance

Most people

On average, people perform surprisingly poorly on this task — often close to guessing.

This highlights how convincing modern AI-generated faces have become.

Try it yourself

See how well you can detect AI-generated faces.

The test is short, free, and designed to reveal how difficult this task has become.