Wisdom on Exaggerative Capabilities of LLMS (And Update on Personal Project Part 1)

Hey, it’s me from the future, and yep, exactly like expected, everything is exaggerated and bs. Like all these random numbers, all these random "1 trillion parameter model," whatever bs...and the model is literally the exact same.  You can’t tell a difference.  It’s very subtle bs differences.  Like, say Claude Opus 4.4 vs 4.5 vs 4.6....they all pretty much feel exactly the same.  Like, you might as well use the newer versions...but like, it’s not like going from Xbox OG to Xbox 360….it’s all pretty much the exact same....