At the core, this question wants to explore how defensible Custom GPTs really are.
Background
OpenAI recently announced plans to release a GPT store, where users can discover and install Custom GPTs created by third-parties. I posted the question below exploring how much revenue a Custom GPT would generate in a single month.
@VAPOR commented saying that "gpts are insecure and can be robbed of their ideas, files, prompts etc and replicated on a cheap third party service" which led to me discussing this claim with @firstuserhere
Resolution Criteria
1. Assumption of Leaderboard: We're assuming OpenAI will have some sort of leaderboard in their GPT store to help people find Custom GPTs. We don't know the exact details yet, but it's a safe bet there will be a way to see which GPTs are popular.
2. Which Chart to Use:
- If there's a global chart, that's what we'll go by.
- If there's no global chart, we'll need to adjust our approach.
- If there are multiple global charts, we'll use the one that shows the number of installs or the number of active users. If both metrics are available, we'll prioritize active users since that's more about how much people are actually using the GPTs.
3. Time Frame and Ranking: The bet focuses on whether any Custom GPT will rank in the top 10 on OpenAI's chart for two consecutive months during 2024.
Related
From: https://twitter.com/NickADobos/status/1729560574768410828?t=g8bN3AUZ7DeHwUFcHVujUw&s=19
The 'GPT builders' update email I got recently implied this issue is when code interpreter is switched on? They disabled it by default now? They might be a little more secure now??