7 In all strategies but hybrid brainstorming we generate a set of 100 ideas in a single GPT session using a specific prompt. In total, for each strategy, we generate 10x100 ideas over 10 sessions to account for temperature and other possible variance. This results in 1000 ideas for each strategy. For hybrid brainstorming, we want to take advantage of multiple “participants” generating ideas and then picking the best. To facilitate this, we initially generate 40x30 ideas in separate sessions, mimicking 40 people individually generating 30 ideas each. Then, we “team up” 4 individual brainstorming sessions into 10 groups of 4 each. Each member of a group brings their 30 ideas “to the group meeting” yielding 120 ideas in total per group. Each group then decides on the best 10 ideas in one session. The results from each group are put together into one pool of 100 ideas. Conceptually, our prompting strategies can be categorized into 8 groups (the full list of all prompts can be found in Appendix D): 1. Idea Prompted GPT: Prompted with seven successful ideas from prior research (same pool as in Girotra et al 2023) 2. HBR trained GPT: Prompt includes a paper on best practices in brainstorming published in the Harvard Business Review (https://hbr.org/2018/03/better-brainstorming) 3. Design Thinking GPT: Prompt includes an article on design thinking by Stanford (https://web.stanford.edu/~mshanks/MichaelShanks/files/509554.pdf) 4. Special Techniques: Involves prompts that offer tipping, plead with the model emotionally, and threaten to shut it off amongst others. 5. Persona Modifiers: Starting with a base prompt, these pools tell the model they are widely known personas, such as Steve Jobs or Sam Altman. It further includes less concrete personas, such as “extremely creative entrepreneurs” adding modifiers such as asking for ideas to be “good”, then “bold”, then “diverse and bold. 6. Hybrid brainstorming: Pool consists of ideas combined from 40 different GPT-4 sessions, picking and refining the best ideas from 4 of them at a time as explained above 7. Similarity information: Informed GPT-4 about the cosine similarity of existing ideas. 8. Chain-of-Thought: Asking GPT-4 to first generate a short list of 100 ideas, then making them bold and different, then generating descriptions for them.