Mode
Both axes. Criterion is met only when both clear.
Targets
0/4000
Forge upgrades the intent into a dual-axis prompt. You can also run the intent directly.
0
SI index
Aim at the criterion
Super-intelligence here is not a claim. It is a score. A run meets the Foltz criterion when computational success and knowledge both clear 85 and the geometric SI index clears 88. Grok forges the prompt, then scores the work against both axes.
Computational0Knowledge0
How a run works
- 1. Intent. Name a problem that can be solved or known — a proof, a protocol, a synthesis with gaps.
- 2. Forge. Grok rewrites it into an SI-grade prompt with success criteria and checks.
- 3. Run. The model works the prompt, then scores itself on computational success and knowledge. Close remaining gaps until the criterion holds.