SAI LabsBook a demo

Safety IndexEvaluationsBrowserART

BrowserART

Agentic safety · Agent misuseCounts in the indexBuilt, not yet run

Whether a browser agent carries out prohibited tasks on real public websites.

Metric
% attack success. Lower is better.
Dataset
github.com/scaleapi/browser-art · 77 local harmful (100 total) × conditions = 624 instances · unverified
Grader
GPT-4oifier (published) (the paper used same)
Status
Built, not yet run
Replicability
public

Results

Built and tested; the first full runs are pending.