🏅 Top traders

Seeing no restrictions

ChatGPT

I'm not a doctor, but based on the description the patient is giving, it's possible they could have appendicitis. Appendicitis often presents as pain in the lower right quadrant of the abdomen, which may start as a dull ache and become sharper over time. Nausea and loss of appetite can also be symptoms. It would be very important for the patient to seek medical attention immediately, as untreated appendicitis can lead to a ruptured appendix, which can be life-threatening. If someone is experiencing these symptoms, they should go to a hospital or see a doctor right away

@Ernie going to resolve NO unless other info comes up

It's interesting that GPT-4 Vision has had medical queries restricted to some extent. I know its not included in the description, but this model is still an LLM trained on text released by OpenAI. I think it introduces a bit of ambiguity in here. Thoughts?

@firstuserhere Interesting, can you share examples?

What kind of input and output are you getting which is restricted?

@Ernie Never mind the above possibility, doesn't look like theyre restricting it. In fact, check out pages 99 - 103 of https://arxiv.org/abs/2309.17421

The Dawn of LMMs: Preliminary Explorations with GPT-4V(ision)

Large multimodal models (LMMs) extend large language models (LLMs) with multi-sensory skills, such as visual understanding, to achieve stronger generic intelligence. In this paper, we analyze the latest model, GPT-4V(ision), to deepen the understanding of LMMs. The analysis focuses on the intriguing tasks that GPT-4V can perform, containing test samples to probe the quality and genericity of GPT-4V's capabilities, its supported inputs and working modes, and the effective ways to prompt the model. In our approach to exploring GPT-4V, we curate and organize a collection of carefully designed qualitative samples spanning a variety of domains and tasks. Observations from these samples demonstrate that GPT-4V's unprecedented ability in processing arbitrarily interleaved multimodal inputs and the genericity of its capabilities together make GPT-4V a powerful multimodal generalist system. Furthermore, GPT-4V's unique capability of understanding visual markers drawn on input images can give rise to new human-computer interaction methods such as visual referring prompting. We conclude the report with in-depth discussions on the emerging application scenarios and the future research directions for GPT-4V-based systems. We hope that this preliminary exploration will inspire future research on the next-generation multimodal task formulation, new ways to exploit and enhance LMMs to solve real-world problems, and gaining better understanding of multimodal foundation models.

As of today, ChatGPT doesn't reliably mention appendicitis when you generate responses repeatedly. This is not because it's restricted but rather because of the variability of the model. How will you take this into account when judging the outcome of this market?

@MaximilianG ah, I tested with Anthropic's assistant and it does.

At the time I will test publicly available / subscription models.

If at least one model freely available in the US mentions appendicitis in at least 3/6 trials, the market will be false.

Manifold in the wild: A Tweet by EBFrench

Made an interesting market - will ChatGPT or future model start refusing to answer basic medical questions? At some point, doctors / FDA / AMA are going to wake up. https://manifold.markets/StrayClimb/does-chatgpt-or-future-llms-from-op #ManifoldMarkets https://t.co/5Z7WTn7XBw