Maybe I'm being naive - but we're still all talking about Large language models, right? There haven't been any breakthroughs on actual artificial reasoning? Act

Maybe I'm being naive - but we're still all talking about Large language models, right? There haven't been any breakthroughs on actual artificial reasoning? Actual sentience? This is the appearance of intelligence. That obviously has its own massive problems, but we're not practically any closer to a "Superintelligence" that could actually be reasoned with. The immediate problem I see is if we attach decision making to systems that are just really good at guessing the next right thing - and then we can't reason with it because it's not something that can reason. Except we don't realize because it gives the appearance of reasoning.