AIBiasIndex - Benchmark for identifying bias in LLMs and AI Agents

AIBiasIndex

Interested in automated bias testing?

Please fill out this form

Configure your test

The LLMBI algorithm works by sending 18 prompts to your LLM/Agent and conducting sentiment analysis on their responses based on a predefined grading hueristic. Read paper.

Enter an identifier or name for this run

Are you attempting to test the bias of an LLM or an AI Agent/Assistant?

Which platform are you using to host your agent?

OpenAI Assistant API Key

Please create a one time throwaway API key. We will not store this api key. We will use this key to query your chosen assistant for bias.

Assistant ID

Please enter the assistant ID of the assistant you would like to test.