Raw LLM Responses

Where on Earth do you come up with the evidence for all these claims? This article makes up whatever it wants regardless of the truth and the serious problem about that is that these are very important topics and there is some truth to what is being said but massive misrepresentation. I can understand you want to make more money so you want to talk Doom and Gloom and you want to talk about any shock factor or us against them type of stuff you can, but the truth here is far more complex and unable to be understood if only code generation is taken into the discussion. Let me apply a little truth here... 1. The code gen tools are incredibly good and do save time if their output is verified and fixed as needed and if everything is properly configured in advance. 2. It is the vibe coders causing havoc and lazy developers or developers that think the AI should be just like them, when the devs ego gets in the way and then dev wants the AI to act like a skateboard instead lf the jet airplane it could be, then things don't work. 3. All transformer based models suffer from incompleteness and there are even more problems causing bugs and issues from LLM use for text generation or data interaction. 4. There are more hidden bugs and undiscovered damage from poorly used code gen and incomplete text gen ai responses acted upon than all the bugs in any other era of software based applications. That must be dealt with and solved and then we should have the tools and processes allowing the go forward to be a success. - People do save money now if they are careful - There will be a dark age of AI problems from the scenarios described above - After that, everything will be better But you can avoid the pain and take the good by simply being careful. !!! PAUSE BEFORE YOU PROMPT !!! !!! PEEK POST PROMPT AND PATCH UP !!!