From my use of Bard, I think it’s main issue is poor alignment due to lack of RLHF dataset. Open AI has been curating its data set for years as it aggressively

From my use of Bard, I think it’s main issue is poor alignment due to lack of RLHF dataset. Open AI has been curating its data set for years as it aggressively pushed to productization. Google never cared about getting its models into the hands of the public so is having to scramble. I think Google will catch up eventually, but not before doing major damage to its market share and partnerships.