FAR AI

non-profit

https://far.ai/

AlignmentResearch

Activity Feed Request to join this org

AI & ML interests

Frontier alignment research to ensure the safe development and deployment of advanced AI systems.

Recent Activity

afspies-far updated a collection 6 days ago

afspies-far updated a collection 6 days ago

afspies-far updated a collection 6 days ago

View all activity

Papers

Exposing the Systematic Vulnerability of Open-Weight Models to Prefill Attacks

View all Papers

updated a collection 6 days ago

Telling FIBS

FIBS Dataset and Probes • 3 items • Updated 6 days ago • 2

updated a dataset 9 days ago

AlignmentResearch/diverse-deception

Viewer • Updated 9 days ago • 11.4k • 222

published a dataset 9 days ago

AlignmentResearch/diverse-deception

Viewer • Updated 9 days ago • 11.4k • 222

updated a model 11 days ago

AlignmentResearch/probe-inference-weights

Updated 11 days ago • 1

updated a model 12 days ago

AlignmentResearch/probe-inference-weights

Updated 11 days ago • 1

updated a dataset 12 days ago

AlignmentResearch/fibs-v1

Viewer • Updated 12 days ago • 163k • 270 • 1

published a dataset 12 days ago

AlignmentResearch/fibs-v1

Viewer • Updated 12 days ago • 163k • 270 • 1

updated a model 13 days ago

AlignmentResearch/probe-inference-weights

Updated 11 days ago • 1

updated a dataset 18 days ago

AlignmentResearch/fibs-v1

Viewer • Updated 12 days ago • 163k • 270 • 1

published a model 19 days ago

AlignmentResearch/probe-inference-weights

Updated 11 days ago • 1

published a dataset 19 days ago

AlignmentResearch/probe-inference-parity

Updated 19 days ago • 158

updated a dataset 19 days ago

AlignmentResearch/probe-inference-parity

Updated 19 days ago • 158

updated a dataset 20 days ago

AlignmentResearch/impossible-swe-nemotron-rollouts

Viewer • Updated 20 days ago • 82.1k • 102 • 2

published a dataset 22 days ago

AlignmentResearch/impossible-swe-nemotron-rollouts

Viewer • Updated 20 days ago • 82.1k • 102 • 2

updated a dataset about 1 month ago

AlignmentResearch/impossible-swegym

Viewer • Updated Sep 15 • 6.86k • 51

published a dataset about 1 month ago

AlignmentResearch/impossible-swegym

Viewer • Updated Sep 15 • 6.86k • 51

submitted a paper to Daily Papers 8 months ago

Exposing the Systematic Vulnerability of Open-Weight Models to Prefill Attacks

Paper • 2602.14689 • Published Feb 16 • 1

authored a paper about 1 year ago

Finding Dori: Memorization in Text-to-Image Diffusion Models Is Less Local Than Assumed

Paper • 2507.16880 • Published Jul 22, 2025 • 7