3 ms·
> wrong metric, it should be an ROI This is true, and I agree. I use GPT-4-0125-preview and Claude 3 Opus for very specific use cases for my work, every single
by consumer451 2y ago
> wrong metric, it should be an ROI
This is true, and I agree. I use GPT-4-0125-preview and Claude 3 Opus for very specific use cases for my work, every single day. This somewhat temperamental AI is a huge economic multiplier for my work. I know the proper use cases, and I get amazing ROI.
> Are my coworkers right X%?
This is a really important point to raise. Just as we don't have a baseline definition for general intelligence, what are the accuracy baselines for customer service rep, legal advisor, medical advisor, etc?