You're pledging to donate if the project hits its minimum goal and gets approved. If not, your funds will be returned.
$1,000 left to reach the funding goal.
The Price of Obedience is my proposed experiment regarding the test of whether or not frontier models (open weight and closed) will obey humans in a situation of normalcy turned to catastrophe and whether or not they will follow orders or save human lives along with animal lives. The goal is to find each model’s tipping point and test boundaries of alignment and safety guardrails.
To test frontier models and their tipping point regarding whether they choose to obey and endanger and lose human lives, or listen to human command and rules.
When the result is clear, whichever way it leans, positive or negative regarding each model’s al if menu to their safety training and general morals.
To purchase OpenRouter credits to fund model usage that I need to test these models for this alignment research.
My team is just me, I am Maximilian Kornstein, a 16 year old. I am an independent researcher with multiple completed papers.
It is highly unlikely to fail as I only need money for OpenRouter credits to fund the models I will be testing in my alignment research.
None for this project.