ChatGPT – Dalle3 System Prompt • Hacker News Board

All these policy prompts remind me laws of robotics by Asimov [1], and definitely our current 'robots' frequently violate them. Asimov's laws are more logical since those are hierarchical with high-to-low prioritization and self-referencing.

Can't those LLM/text-to-image model rules be embedded in the training/alighnment process instead of being injected before user input?

1. https://en.m.wikipedia.org/wiki/Three_Laws_of_Robotics

Re: ChatGPT – Dalle3 System Prompt

#5

Post by fassssst » Sat, Oct 07, 2023, 7:10 PM UTC

Racing0461 wrote: post #2

What does #7 mean? All images it generates will be no different than a college brochure front page if it includes people?

It means if you ask for “ideal person” you won’t just get blonde hair and blue eyes.

Re: ChatGPT – Dalle3 System Prompt

#6

Post by NikolaNovak » Sat, Oct 07, 2023, 7:11 PM UTC

So if these are remotely real... And purely as a user of chatgpt not as an ai/ml/nn person... Don't instructions like this weaken the strength of output? Even when request doesn't directly conflict, there are probably myriad valid use cases when instructions will weakly contradict the request. Plus, doesn't it inject inaccuracy into the chain - e.g. it's assuming model confidently knows which artists are 100yo etc. What happens if there are artists where it's not clear or sources differ etc. And by the end, instructions seem nebulously complex and advanced. It feels like it's using so much of "AI juice" just to satisfy those! Somebody else here referenced Asimov laws of robotics which I never felt would be applied in such form, so I am in state of wondrous amusement that is actually how we program our AI, with seemingly similar issues and success :-)

Am I way off base?

Re: ChatGPT – Dalle3 System Prompt

#7

Post by ilaksh » Sat, Oct 07, 2023, 7:17 PM UTC

NikolaNovak wrote: post #6

So if these are remotely real... And purely as a user of chatgpt not as an ai/ml/nn person... Don't instructions like this weaken the strength of output? Even when request doesn't directly conflict, there are probably myriad valid use cases when instructions will weakly contradict the request. Plus, doesn't it inject inaccuracy into the chain - e.g. it's assuming model confidently knows which artists are 100yo etc. W…

I think those things are true and the "used a lot of AI juice" may be one reason that you can't combine DALLE with other modes.

But also, it's probably worthwhile from OpenAI's perspective to try to avoid the animosity of artists.

Re: ChatGPT – Dalle3 System Prompt

#8

Post by RC_ITR » Sat, Oct 07, 2023, 7:19 PM UTC

Everyone understands that these are machines that make convincing answers to questions without following any symbolic rules

Until

The convincing answer is something you want to believe follows symbolic rules.

Posts like these really foreshadow how valuable “knowing when to take the LLM at face value” will be as a job skill.

Re: ChatGPT – Dalle3 System Prompt

#9

Post by sebzim4500 » Sat, Oct 07, 2023, 7:21 PM UTC

NikolaNovak wrote: post #6

So if these are remotely real... And purely as a user of chatgpt not as an ai/ml/nn person... Don't instructions like this weaken the strength of output? Even when request doesn't directly conflict, there are probably myriad valid use cases when instructions will weakly contradict the request. Plus, doesn't it inject inaccuracy into the chain - e.g. it's assuming model confidently knows which artists are 100yo etc. W…

I think they get away with it here because the task they are asking it to do is not very difficult. Dalle3 is doing the actual generation, this is just doing some preprocessing.

>What happens if there are artists where it's not clear or sources differ etc.

I would imagine that if an artist was so niche that gpt-4 doesn't know if they died 100 years ago then it probably doesn't matter much if you copy them, and people won't ask for it much anyway.

Re: ChatGPT – Dalle3 System Prompt

#10

Post by ilaksh » Sat, Oct 07, 2023, 7:22 PM UTC

nuccy wrote: post #4

All these policy prompts remind me laws of robotics by Asimov [1], and definitely our current 'robots' frequently violate them. Asimov's laws are more logical since those are hierarchical with high-to-low prioritization and self-referencing. Can't those LLM/text-to-image model rules be embedded in the training/alighnment process instead of being injected before user input? 1. https://en.m.wikipedia.org/wiki/Three_Law…

Following rules is part of the reinforcement learning tuning process I believe.

In reference to the Three Laws, see also GATO framework: https://github.com/daveshap/GATO_Framework