I've been noticing something strange lately...
People rarely ask AI if it's intelligent.
They ask if it's right.
That's a completely different question u know.
The AI industry seems obsessed with making models more capable.
But most users are trying to solve a much simpler problem...
Can I trust this output???
That's why I think AI's biggest challenge isn't intelligence. It's confidence.
Not confidence in the model.
Confidence in the answer.
Where did it come from?? Can it be verified??
Can someone else reach the same conclusion??
That's what makes projects like #OpenGradient interesting to me.
Maybe the future of AI won't be decided by who generates the most answers.
Maybe it'll be decided by who makes answers easier to trust.
Just sharing a thought...
What's your view?