Pick the Winner: results
Export CSV Export JSON Back to the game
Experts
Each expert vs the models on the same pairs, in the order that expert saw. Level: mean confidence, 1 Not sure to 4 Certain.
| Name | Role | Pairs | Expert | Opus 5.5 | GPT-6.1 Sol | Level | Median s |
|---|
Pairs
Experts right / experts who played it. The models were asked both orders (0 to 2 right).
| Pair | Half | Experts right | Opus right | GPT right |
|---|
Invites
Add emails (one per line or comma-separated). No email is sent: tell people to open humans.convictionlabs.ai and sign in. @convictionlabs.ai and @bharatmavens.com addresses can always sign in.
| Name | Role | Played | Invited |
|---|