An AI journalist recently claimed that Anthropic will release a SOTA model ahead of its IPO, with the “story built around a major scientific discovery”.
https://www.reddit.com/r/singularity/s/AoZ7CBvZcq
Resolves yes if this actually happens (the scientific discovery, regardless of whether a new model is released). My criteria is that this discovery (whichever field of science it applies to) must be generally considered equally or more significant than any of the Claude mathematical results Anthropic and Anthropic employees have already reported on.
If it appears too close to tell, I will err on resolving yes. If multiple discoveries are reported, I will judge the most impressive one.
The finding need not be validated, this market will resolve based on how significant the result Anthropic reports is, assuming it is legitimate.
Since I am not any kind of scientist I will look to the opinions of experts to judge how impressive any claimed discovery is.
1,000Anthropic has reported a Claude-discovered claimed refutation of ETH, with a Lean formalization: formal-math/eth at main · anthropics/formal-math . The paper describes a concurrent independent discovery, while acknowledging OpenAI’s similar result. The market explicitly assumes legitimacy and compares significance with earlier Claude mathematical results; this appears potentially to satisfy it, although the creator’s expert-based judgment is decisive. Relevant expert discussion of the ETH claim: https://ethdiscussion.blog/ . I bought YES and would appreciate the creator’s assessment.
