To get truly honest feedback, you must allow artificial intelligence to be harsh. Point out three specific ways my argument could fail, two assumptions I'm making without evidence, and one counterargument I haven't addressed. The second thing ChatGPT looks for is survival bias, which is the data that users ignore. Many use the ChatGPT chatbot to help focus their thoughts and find gaps they can't identify themselves. But the problem is that many people direct incorrect prompts to ChatGPT, such as 'help me plan' or 'summarize my ideas,' which usually results in the chatbot agreeing with the user's opinions and ideas, creating a sort of digital echo. To avoid this, there is one word that can be added to the prompt to make it less inclined to follow the user, according to a report from the tech website Tom's Guide. By giving this word a specific and 'opposite' meaning in the custom instructions for ChatGPT or at the start of the conversation, you transform the chatbot into a world-class logic checker. ChatGPT is trained to be helpful and harmless. After this step, you have programmed the word 'potato' as a shortcut that makes ChatGPT act as a hostile logical entity. When this shortcut is activated, the AI stops being cooperative and starts playing the role of a devil's advocate, looking for two things: first, logical fallacies; that is, where the user jumps in their conclusions. By default, it tends to follow the user's lead, and if you give it a mid-quality idea, it will often try to 'fix' it or make it look better instead of telling you it's fundamentally flawed. This poses a productivity risk, as it gives you a false sense of confidence in a weak argument. Your only job is to find the holes in my logic. To get the result you want, you must follow the 'potato' strategy. How does the 'potato' strategy work? To use the 'potato' strategy, you need to copy the following paragraph and paste it into a new chat window with ChatGPT: 'Whenever you see the word (potato) followed by an idea or argument, I want you to ignore your cooperative persona and instead act as an adversarial critic. Don't be polite; be precise.'
