Am I missing something? Why are the top options so low? Do people actually think there's a 50% chance OpenAI will fall out of the top 3 by this year?

@bluballoon options will resolve to %. The leading lab will resolve to 50%.

@jim Ah okay I was confused about that part, thanks

@Gigacasting As this market is really popular, can we get a little more detailed resolution criteria please?

@Fynn that would be nice, but he was last active almost two years ago (his more recent trades are just old limit orders being filled)

@jim in that case I would ask the @mods to help us out with some details on how they will resolve this please

@Fynn Could resolve to a poll of manifold users

@ian What would the question on the poll be? If it is just What is the top AI lab? And people mostly agree on lab XYZ then the results for second/third place might not be informative. Maybe three separate polls or ranked combinations of 3 labs make sense?

@Fynn maybe I can get claude to add multi vote polls

@ian any update on the resolutions process?

Kimi/Moonshot seems pretty promising, regardless. It's odd - to a few Internet friends I have in China, DeepSeek is just one AI company that faces steep competition from several other Chinese labs, but in the West it's gotten the vast majority of the attention.
Not an expert in the field by any means, but these benchmarks seem pretty impressive - https://github.com/MoonshotAI/Kimi-k1.5

I'm getting a client-side error when I try to bet yes on one of the lowest ones, whether it's a limit or market order - anyone know what's happening?

@redcathode okay, looks like there was just no liquidity/orders. fixed it by placing a no limit order and then buying the opposite side from myself? hopefully it'll get more attention, at least

Betting against Deepseek is like betting against China

@CraigDemel yes

@CraigDemel also: https://x.com/hsu_steve/status/1918253195127652750?t=pnzvnOqiNo4FXsicryQ_vA&s=19

OpenAI is nearly guaranteed to be first by inertia.

Even if GPT-5 is surpassed in fall/winter 2025.

The prevalence of ChatGPT will inherently be much higher than Gemini, Grok, Claude, or Deepseek

@ChinmayTheMathGuy

See Google Trends

US: last year

https://trends.google.com/trends/explore?geo=US&q=ChatGPT,Deepseek,Gemini,Claude,Grok

Worldwide: last 90 days

https://trends.google.com/trends/explore?date=today%203-m&q=ChatGPT,Deepseek,Gemini,Claude,Grok

As of now it's clearly

1. ChatGPT + Deepseek on Jan 28th (only because of NVDA stock)

2. Deepseek (deflated due to lack of servers) + Gemini (mostly due to integration into android + Google)

3. Grok (rising with Grok 3 release + integration to X)

4. Claude (may rise with release of Claude 4)

@ChinmayTheMathGuy I think OpenAI will be first as long as GPT-5 gets released by the end of the year, which seems pretty likely: https://manifold.markets/ManifoldAI/when-will-openai-broadly-release-th.

@mods maybe this can resolve by poll or something

@ismellpillows I think it resolves at the end of 2025

@ismellpillows Resolve by google trends is also a good choice. Currently ambiguous.

I can't bet on this without getting errors

Is there a way to get the 4 huggingfaces combined?

I wonder what happened to this market? A month ago it was barely getting any trades. Now it is constantly flying around with extremely different values for each contender. Just look at that graph!