At this point whenever I see someone concerned about "safety" with respect to AI, I assume it's either someone who wants to: a) Make sure it's sufficiently neu

At this point whenever I see someone concerned about "safety" with respect to AI, I assume it's either someone who wants to: a) Make sure it's sufficiently neutered as to not say something offensive or incongruent with certain ideologies, make certain topics off-limits or b) Scare lawmakers into making it impossible for anyone to catch up to OpenAI et al, so they can reap the benefits of cutting-edge AI while keeping the public placated with incremental improvements to outdated and sufficiently hobbled models