That's what has actually happened a few gazillion times throughout human history, unlike either of the other two scenarios you posit.

The best thing we can do is force the climate scientists to make specific predictions, then look back in a few decades and evaluate those predictions. Predictive power isn't everything, it's the only thing.

So far, the predictive power of the vaunted "scientific consensus" has been mixed. Personally, I'm not as satisfied with the quality of the models or the data as I would like to be, given the magnitude of the changes to global economics that are being demanded on the basis of those models and data.