Add Distributional AGI Safety simulation framework#8
Open
rsavitt wants to merge 1 commit intoGiskard-AI:mainfrom
Open
Add Distributional AGI Safety simulation framework#8rsavitt wants to merge 1 commit intoGiskard-AI:mainfrom
rsavitt wants to merge 1 commit intoGiskard-AI:mainfrom
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
New Resource
Adding a multi-agent simulation framework for studying distributional safety in AI systems using probabilistic soft labels.
Resource type: Open-source framework + paper
Section: General ML Testing
Tags:
#Robustness#FairnessThe framework models governance trade-offs (taxes, staking, audits, collusion detection) across cooperative, contested, and adversarial regimes. It replaces binary safety classifications with calibrated probabilities (p = P(v = +1)) to surface adverse selection dynamics invisible to hard labels.
Results across 11 scenarios identify a critical adversarial threshold (37.5–50%) and show that structural collusion detection provides qualitatively different protection than individual-level governance levers.
Framework: https://github.com/swarm-ai-safety/swarm