{"type":"post","id":"new-open-source-asimov-jailbreak-benchmark-tests-adversarial-38a3a7","slug":"new-open-source-asimov-jailbreak-benchmark-tests-adversarial-38a3a7","title":"New open-source ASIMOV-Jailbreak benchmark tests adversarial robotics prompts; Argon scores 99.5%.","description":"Google has published ASIMOV Agentic, an open-source evaluation harness and dataset for testing AI systems on safety-critical robotics tasks.","retrieval_nugget":"Google has published ASIMOV Agentic, an open-source evaluation harness and dataset for testing AI systems on safety-critical robotics tasks. It gives teams concrete tests for tool use, perception, physical constraints, and multi-turn jailbreaks, but benchmark results alone do not establish that a robot is safe in deployment.","published_at":"2026-11-10","updated_at":"2026-11-10","record_date":"2026-11-10","date_kind":"published_at","topics":["google"],"entities":["Google","ASIMOV Agentic","Hugging Face","Inspect AI"],"related_patterns":["goal-scoped-loops-replace-manual-continuation"],"novelty_label":"new_fact","what_changed":"A new open-source benchmark tests adversarial prompts for robotics, with Gemini 4 Argon reportedly scoring 99.5%.","prior_refs":[],"source_url":"https://huggingface.co/datasets/google/asimov_agentic","source_urls":["https://huggingface.co/datasets/google/asimov_agentic"],"schema_version":"newruntime-agent-readable-v0.2","stable_id":"post:new-open-source-asimov-jailbreak-benchmark-tests-adversarial-38a3a7","status":"published","visuals":[],"editor":{"role":"editor","name":"Andrey Reshetnikov","profile":"https://newruntime.com/owner-profile.md","profile_json":"https://newruntime.com/owner-profile.json","same_as":["https://www.linkedin.com/in/reshetnikov1/","https://t.me/lapetuse"]},"routes":{"html":"https://newruntime.com/posts/new-open-source-asimov-jailbreak-benchmark-tests-adversarial-38a3a7/","markdown":"https://newruntime.com/posts/new-open-source-asimov-jailbreak-benchmark-tests-adversarial-38a3a7.md","json":"https://newruntime.com/posts/new-open-source-asimov-jailbreak-benchmark-tests-adversarial-38a3a7.json"},"markdown_url":"https://newruntime.com/posts/new-open-source-asimov-jailbreak-benchmark-tests-adversarial-38a3a7.md","next_reads":[{"type":"topic","path":"/topics/google/","url":"https://newruntime.com/topics/google/","title":"Google","reason":"Explore the google topic hub.","media_type":"text/html","markdown_url":"https://newruntime.com/topics/google.md","json_url":"https://newruntime.com/topics/google.json"},{"type":"collection","path":"/posts/","url":"https://newruntime.com/posts/","title":"Field Notes","reason":"Open the latest editorial analysis.","media_type":"text/html","json_url":"https://newruntime.com/posts.json"},{"type":"collection","path":"/sources/","url":"https://newruntime.com/sources/","title":"Sources","reason":"See the original sources this archive cites.","media_type":"text/html"}]}