What does being "intense" on the safety side mean? High risk taking with AI safety or low?
Are you trying to downplay the compute required to train and run inference on these large language models by stringing together some contrived comparison to the now 'uncool' Blockchain technology? That would be absurd.