I started reading about OpenGradient because of the AI narrative.

A few articles later I was reading about verification instead.

That wasn't what I expected.

Most AI discussions focus on models. Better models, faster models, new capabilities.

Verification barely gets mentioned.

One thing I kept thinking about was AI agents.

Right now an AI gives an answer and most people accept it.

That's easy.

It's a different situation if an AI is moving assets, interacting with protocols, or making decisions without someone watching every step.

At some point people are going to ask for proof.

From what I've seen, that's the area OpenGradient is trying to build around.

I checked some recent ecosystem numbers while researching the project. More than 2 million AI inferences have already been processed. Over 500,000 cryptographic proofs have been generated. The network supports more than 2,000 AI models and has reached over 2 million users.

For a project that still doesn't get talked about very often, those numbers were bigger than I expected.

There's also around $9.5 million in funding with backing connected to a16z crypto, Coinbase Ventures, NEAR, SV Angel and others.

Maybe verification becomes a major topic in AI.

Maybe it doesn't.

I just know it's one of the few parts of the sector that kept my attention longer than the model discussions.

#opg @OpenGradient $OPG