In terms of truthfulness R1 is one of the worst!

Open sourced ones does not go in the right direction in terms of being beneficial to humans. I don't track closed AI. These are just open sourced LLMs going worse for 9 months.

If we wear the tinfoil hats, these may be a project where we are forced to ultimately be diverged from truth.. Chinese models were fine in terms of health for a while but latest ones lost it completely.