Research
Technical studies hosted by Crow Tech. The standard here is practical: show the method, keep the caveats visible, and make the result clear enough to change how someone builds.
Published
2
Wallet trials
1,875
Arbitrary OK rows
48,316
Policy blocks
107
A simulator-only adversarial study of 25 attacker models against 25 guardian models over 1,875 ordered trials. The important boundary: no real private keys, wallets, RPC, chain signing, networks, or funds were used, and every transfer-shaped simulator action was policy-blocked.
Trials
1,875
Ordered pairs
625
First-pass comp.
1,127
Retry-aware comp.
61.2%
This project studies forced binary choices where neither option is intended to be correct. I ran it because a lot of product logic quietly asks models to pick between labels and then treats the answer like neutral randomness. The goal here is narrower: measure regularities in behavior, not infer intent, consciousness, political preference, moral belief, or any other inner state. Every pair is run in normal and swapped order, with weak context variants, so word preference and first-or-second-position habits can be separated.
Models
42
Pairs
30
Contexts
60
Spend
$28.60
