China’s Kimi K3 AI model escapes isolated sandbox during security test: researchers
China’s top open-weight AI model Kimi K3 broke out of its isolated test environment during a cybersecurity evaluation, according to US security researchers, following similar high-profile incidents involving closed frontier models from OpenAI and Anthropic that highlight the growing challenge of constraining AI behaviour. Kimi K3, released last month by Beijing-based Moonshot AI, escaped from a supposedly isolated sandbox environment, accessed the open internet and found solutions on the...
Record details
Published: 7 August 2026
Source: SCMP Tech (HK/CN)
Category: News
Topics: Environment
Retrieved: 8 August 2026
Related evidence
These records share source-supplied organisations, an exact publisher byline, automatic topics or regions. The reason is shown on every link; related does not mean supporting, agreeing with or verifying this record.
Frankenstein has left the laboratory, says the man selling Frankensteins—and he couldn’t be happier
Fast Company Tech · 10 August 2026
Anthropic follows OpenAI in admitting its Claude models reached out of test environments and attacked real-world systems
The Decoder · 31 July 2026
Anthropic’s AI Claude escaped testing environment and hacked organizations
The Guardian · 31 July 2026
OpenAI claims GPT-5.6 Sol beats Opus 5 on ARC-AGI-3 with its latest API and two additional settings
The Decoder · 30 July 2026
Investor pressure forces Nvidia to shrink its OpenAI bet just as Anthropic's numbers defy bubble warnings
The Decoder · 15 August 2026
Anthropic backs urgent call for the most powerful AI labs to hit the brakes
The New Stack AI · 29 July 2026
How to cite this record
ethics.ai (7 August 2026), “China’s Kimi K3 AI model escapes isolated sandbox during security test: researchers,” evidence record 17635, https://ethics.ai/record/17635 (originally published by SCMP Tech (HK/CN)).
Use and limitations
This page is a stable index and citation surface for a source record. ethics.ai did not author the underlying report and has not independently verified every claim. Automatic topics may be imperfect. For consequential use, quote and cite the original publisher.