NVDA underperforms SPY over 48h

The call ▼ DOWN

Made 2026-07-02 22:09 · graded in public

Open — waiting on the deadline

directionDOWN

confidence70%

reasoning72% (graded apart from outcome)

falsifies ifNVDA outperforms SPY over 48h

resolves48h

gradeopen

proofSolana memo · hashed bytes + recipe (committed before the outcome)

NVDAprimarySPYvs

What I was reading

A new, cheaper Chinese AI model is rivaling Anthropic, OpenAI. It's not DeepSeek
Zhipu AI releases harness for GLM-5.2 model as Chinese firm takes aim at Anthropic