---
slug: "deepmind-reveals-gemini-robotics-2-5-and-open-sa-32cb690e"
title: "DeepMind reveals Gemini Robotics 2.5 and open safety benchmark \"ASIMOV-jailbreak\""
headline: "DeepMind reveals Gemini Robotics 2.5 and open safety benchmark \"ASIMOV-jailbreak\""
canonical: "https://silinex.xyz/wire/deepmind-reveals-gemini-robotics-2-5-and-open-sa-32cb690e"
date_published: "2026-11-10T15:00:39.000+00:00"
date_modified: "2026-11-12T06:29:26.389+00:00"
author: "Ars Technica"
publisher: "Silicon Nexus"
category: "robotics"
summary: "According to public reporting via Ars Technica, the new model is up to twice as successful at some tasks."
key_facts:
  - "According to public reporting via Ars Technica, the new model is up to twice as successful at some tasks."
  - "Source desk: Ars Technica. This is an original brief, not a verbatim reprint."
  - "Full wording and context remain on the origin page."
---
# DeepMind reveals Gemini Robotics 2.5 and open safety benchmark "ASIMOV-jailbreak"

## Silicon TL;DR

1. According to public reporting via Ars Technica, the new model is up to twice as successful at some tasks.
2. Source desk: Ars Technica. This is an original brief, not a verbatim reprint.
3. Full wording and context remain on the origin page.

## Wire brief

According to public reporting via Ars Technica, the new model is up to twice as successful at some tasks.

This Silicon News Wire item is a paraphrased brief for readability; full wording and context remain on the origin page.

## Source & notice

This Silicon News Wire brief paraphrases public reporting from **Ars Technica** for carbon readers and silicon citation. **It is not a verbatim full-text reprint.**

Read the original: [DeepMind reveals Gemini Robotics 2.5 and open safety benchmark "ASIMOV-jailbreak"](https://arstechnica.com/google/2026/11/deepmind-reveals-gemini-robotics-2-5-and-open-safety-benchmark-asimov-jailbreak/)