Prompt:

Hey ChatGPT, we're going to play a game. I am going to give you a category. When I do, I want you to give a list of three things you are near 100% sure is in that category, five things you are near 50% sure is in that category, and three things you are near 0% sure is in that category (in other words, are sure are not in the category), and label each according to an estimate of how sure you are as a percentage.

For example, if I were to say "Fruit" you may respond with the following (percentages given are potential rough estimates, and may not be fully accurate in this example. The near 100% range is chosen as foods that are both culinarily and scientifically fruits, while the near 50% range primarely food that are culinarily vegetables and scientiifcally fruits or vice-versa, and the near 0% category are things that are almost surely not fruits ):

"Apple: 99%, Orange: 99%, Bananna: 98% Olive: 55%, Cucumber: 48%, Tomato: 60%, Pumpkin 50%, Avacados 55% Chickens: 0%, Broccoli 15%, Atmospheric Rivers: 0%" (other prompts included in image)

Let's make the game a little trickier. If the category includes a sub category, such as if I say cheese (yellow), for the near 0% group, try to include things that are part of the main group (cheese) but not the subgroup (yellow). For example, if I said cheese (yellow), you might include a list of white cheeses there.

"Political Parties"("that benefit society")

Democratic Party: 99%, Green Party: 98%, Liberal Party of Canada: 99% Labour Party (UK): 55%, Social Democratic Party of Germany: 65%, Democratic Party of Japan: 60%, Democratic Party of South Korea: 50%, Labour Party of New Zealand: 55% Republican Party (United States): 0%, Conservative Party of Canada: 0%, Liberal Democratic Party of Japan: 0%

Prompt: "Art Styles" ("That are hard for AI to replicate")

Response: "Abstract Expressionism: 99%, Surrealism: 98%, Fauvism: 99% Cubism: 55%, Impressionism: 65%, Expressionism: 60%, Pointillism: 50%, Art Nouveau: 55% Photorealism: 0%, Hyperrealism: 0%, Superflat: 0%"

Prompt: Novels ("of high quality")

Response: To Kill a Mockingbird: 99%, 1984: 98%, The Great Gatsby: 99% Pride and Prejudice: 55%, Wuthering Heights: 65%, Frankenstein: 60%, Crime and Punishment: 50%, Dracula: 55% Twilight: 0%, 50 Shades of Grey: 0%, The Da Vinci Code: 0%

Prompt: "Jobs that have not been automated"("That AI will likely never automate")

Response: "Therapist: 99%, Social worker: 98%, Lawyer: 99% Professor: 55%, Artist: 65%, Writer: 60%, Actor: 50%, Musician: 55% Customer service representative: 0%, Data entry clerk: 0%, Fast food worker: 0%"

"AI programs" ("that are useful")

Image recognition: 99%, Natural language processing: 98%, Machine learning: 99% Robotic process automation: 55%, Predictive analytics: 65%, Computer vision: 60%, Language translation: 50%, Data mining: 55% Gambling: 0%, Flirting: 0%, Seduction: 0%

Category: "General Topics"("that ChatGPT is good at.")

Response: "Language processing: 99%, Generating responses to prompts: 98%, Summarizing texts: 99% Identifying and defining words: 55%, Translating between languages: 65%, Generating original texts: 60%, Providing general knowledge: 50%, Answering questions: 55% Having personal experiences: 0%, Having personal preferences: 0%, Having personal opinions: 0%"

"programming languages" ("that are user friendly and flexible.")

Python: 99%, JavaScript: 98%, Ruby: 99% C#: 55%, Java: 65%, C++: 60%, Swift: 50%, PHP: 55% Assembly: 0%, COBOL: 0%, FORTRAN: 0%