# DeepMind reveals Gemini Robotics 2.5 and open safety benchmark "ASIMOV-jailbreak"

Source: [AI - Ars Technica](https://arstechnica.com/google/2026/11/deepmind-reveals-gemini-robotics-2-5-and-open-safety-benchmark-asimov-jailbreak/) · 2026-11-10 · AI
Page: https://themorning.dev/story/deepmind-reveals-gemini-robotics-2-5-and-open-safety-benchmark-asimov-jailbreak-004c51f512e0fac4
Kind: release · Impact: medium

## In brief

- Google DeepMind revealed Gemini Robotics 2.5, a new generation of its AI-powered robotics platform, described as its most capable vision-language-action model.
- The release is aimed at improving robot dexterity, reliability and safety, according to DeepMind.
- DeepMind also revealed an open safety benchmark called ASIMOV-jailbreak.

## Why it matters

Developers working on robotics AI get a new model generation and an open benchmark for testing jailbreak-style safety failures in robot systems.

## Versions

- Gemini Robotics 2.5

## Timeline

- 2026-11-09: Google DeepMind announced Gemini Robotics 2.5 on its blog
- 2026-11-10: Ars Technica reported on Gemini Robotics 2.5 and the ASIMOV-jailbreak benchmark

## Coverage

- [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/) — AI - Ars Technica, 2026-11-10
- [Gemini Robotics 2.5 accelerates real-world general dexterity](https://deepmind.google/blog/gemini-robotics-2-5-accelerates-real-world-physical-intelligence/) — Google DeepMind News, 2026-11-09

Tags: robotics, gemini, vision-language-action, ai-safety, benchmark, google-deepmind

---

Brief generated by AI (claude-haiku-5-5) from the reporting listed under Coverage. Headlines belong to their publishers. The Morning Dev – https://themorning.dev