As an AI researcher, I have designed an innovation tool that overcomes ChatGPT's weaknesses to innovation (and those of other LLMs). While ChatGPT works with common associations, this new tool works with uncommon associations that are still semantically close. From NSF funded research, it is known that these uncommon yet semantically close associations are the key to creative solutions.Also, as an AI researcher, I have developed a new model in which classic AI, ML, and Humans all collaborate to counter each other's weaknesses in creative problem solving. Called BrainSwarming, it has been popularized by Harvard Business Review videos (see below) as an in-person group problem solving process. It is soon to become an online platform with AI and ML added in.As a computational complexity theorist, I have authored the first mathematical proof that there is a limit to how creative a computer can be. It is a type of Incompleteness Theorem. Published in AI EDAM, it shows that the features of any object are not computably enumerable. As a cognitive psychologist, I note all the mental obstacles to innovation and devise counter techniques. "Overcome Any Obstacle to Creativity," my book that articulates all these techniques, is available on Amazon.com.I also work with math savants to help them clarify their ideas. My first savant's results in number theory will be made public soon.