- You should probably not address the assistant by the second person in the system prompt, for the transcript refers to the assistant as "Assistant", not "You".
- Instead of telling it what to say in the system prompt, you should just prefill the first answer.
Have you considered making this as a custom GPT instead, or as a first step? That way you just need a chatGPT plus subscription and don't need to worry about API costs, and it's extremely easy to set up. Although you wouldn't be able to log usage. And I think from there, it should be pretty simple to connect an API to your custom GPT (and at that point I think more logging should be possible).
I might take a whack at it if I find some time and my mood allows. I reckon it could be done by someone with limited programming experience by consulting with an LLM, but it's complex enough to make that a pretty frustrating and time consuming experience. With the additional guardrails (integration with Google auth, locking by account and IP), it might merit at least an audit by a developer.
What about when someone makes "Pro-War Bot"? Do we then see which is more persuasive? I'm not even going to make the joke about "No War Bot" and "Pro-War Bot" trying to convince each other, it'd probably just loop like a zillion similar discussions.
Come to think of it, isn't that loop basically what Middle East politics is, all the time? Though maybe it's kind of interesting to see if there's a "bot superiority" possible outcome, that one of the bots is better than the other. There's no reason to think they're exactly evenly matched, one of them could very well have better programmers behind it.
This seems trivial, isn't it? You're just calling an existing LLM's API, with no further tooling or customization. You could just ask an LLM to write you the HTML and JS and it can probably do it.
Maybe also have "No-War-Bot" defend the argument that the United States should also not have fought in the Civil War, World War One, World War Two, the first Gulf War, the invasion of Haiti, or the bombing of Kosovo.
- You should probably not address the assistant by the second person in the system prompt, for the transcript refers to the assistant as "Assistant", not "You".
- Instead of telling it what to say in the system prompt, you should just prefill the first answer.
Have you considered making this as a custom GPT instead, or as a first step? That way you just need a chatGPT plus subscription and don't need to worry about API costs, and it's extremely easy to set up. Although you wouldn't be able to log usage. And I think from there, it should be pretty simple to connect an API to your custom GPT (and at that point I think more logging should be possible).
I might take a whack at it if I find some time and my mood allows. I reckon it could be done by someone with limited programming experience by consulting with an LLM, but it's complex enough to make that a pretty frustrating and time consuming experience. With the additional guardrails (integration with Google auth, locking by account and IP), it might merit at least an audit by a developer.
What about when someone makes "Pro-War Bot"? Do we then see which is more persuasive? I'm not even going to make the joke about "No War Bot" and "Pro-War Bot" trying to convince each other, it'd probably just loop like a zillion similar discussions.
Come to think of it, isn't that loop basically what Middle East politics is, all the time? Though maybe it's kind of interesting to see if there's a "bot superiority" possible outcome, that one of the bots is better than the other. There's no reason to think they're exactly evenly matched, one of them could very well have better programmers behind it.
I asked chat gpt and it said you're well intentioned and likely to make a support tool for people who already believe that war is bad.
This seems trivial, isn't it? You're just calling an existing LLM's API, with no further tooling or customization. You could just ask an LLM to write you the HTML and JS and it can probably do it.
Maybe also have "No-War-Bot" defend the argument that the United States should also not have fought in the Civil War, World War One, World War Two, the first Gulf War, the invasion of Haiti, or the bombing of Kosovo.