AI Watermarking Is Not Going to Save Us
A conversation with AI policy executive Irene Solaiman
The Multiple Faces of Claude AI: Different Answers, Same Model
It’s often hard to tell which version of an AI model you’re using — and what, if any, safety settings are in place
How Proof’s AI Coverage Is Inspired by Science
From Reuters
Leading AI Companies Fail to Address Problems With Inaccurate Election Information
AI models continue to provide inaccurate and misleading responses to voter queries
Leading AI Companies OpenAI and Anthropic Are Not Keeping Their Election Promises
The discrepancies between the companies’ public promises — and their execution — raise questions about their commitment to providing accurate information during this high-stakes election year
Seeking Reliable Election Information? Don’t Trust AI
Experts testing five leading AI models found the answers were often inaccurate, misleading, and even downright harmful
How We Tested Leading AI Models Performance on Election Queries
An expert-led domain-specific approach to measuring AI safety
A Letter From Our Founder
At Proof, we don’t chase stories. We develop hypotheses and test them. We aim to be transparent about what we know and what we don’t.