Raw LLM Responses

I work in healthcare and insurance. AI has been a bit of a holy grail for a while but it hasn't delivered. The answer is simple. Complexity of business processes. In my automotive industry days it was the same. "Robots will replace workers" etc. Never happened. Cars weren't designed for manufacturing and assembly (DFM / DFA / DFR). Successful manufacturers always consider those. The rest... Classic example. I did a lot of the above in my PhD work (focused on human factors and UX) and in one immemorabe case the factory i used for a semester project was so bad efficiency wise that we basically concluded that the only way it was designed this badly was that they intentionally knew what not to do but they did it anyway. By contrast, manufacturing operations at a different plant run by a renowned japanese maker were textbook good. Business is like that too. We have coded AI solutions for some things like customer service answers or what not etc the important stuff (enrollment, claims) is too complicated to AI away given current business processes. If the processes are streamlined and documented it's not difficult to AI some of it. And it's not by design, everything in most healthcare and insurance is duct taped together due to mergers and acquisitions so...