Skip to main content
U.S. flag

An official website of the United States government

Official websites use .gov
A .gov website belongs to an official government organization in the United States.

Secure .gov websites use HTTPS
A lock ( ) or https:// means you’ve safely connected to the .gov website. Share sensitive information only on official, secure websites.

Our Projects & Programs

Generative AI Evaluations
GenAI Graphic
Credit: S.Sharma/NIST

The NIST GenAI Program

NIST GenAI is an AI testing and evaluation (T&E) program administered by the NIST Information Technology Laboratory to assess generative AI technologies developed by communities worldwide. It provides rigorous, science-based evaluations of Generators (generative AI), Detectors (discriminative AI), and Prompters (prompt engineering) across multiple modalities (text, code, image, audio, video, multimodal). GenAI employs a dynamic and adversarial testing framework that involves various types of AI actors: Generators create synthetic data (AI content) using frontier AI models, while Detectors develop AI tools to identify whether the content is AI-generated, a Deepfake, or Watermarked. Prompters provide LLMs with inputs describing requirements for AI content to be generated, employing various prompting strategies.

 

Current & Upcoming Challenges 

 

 

Picture for Watermark 2026
Credit: S.Sharma/NIST |
Generative AI was used in the creation of this visual asset. Created on September 16, 2026, by Gemini. More info

 

Picture for Audio 2026
Credit: S.Sharma/NIST |
Generative AI was used in the creation of this visual asset. Created on September 16, 2026, by Gemini. More info
Picture for Code 2026
Credit: S.Sharma/NIST |
Generative AI was used in the creation of this visual asset. Created on September 16, 2026, by Gemini. More info
Picture for Deepfake 2026
Credit: S.Sharma/NIST |
Generative AI was used in the creation of this visual asset. Created on September 16, 2026, by Gemini. More info

Past Challenges

Forensics In AI

Coming Soon!

Contacts

Created July 29, 2016, Updated September 22, 2026
Was this page helpful?