South China Morning Post
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…
View original source — South China Morning Post ↗
Related stories
Euronews
ScienceAug 7, 2026 · 1 min
Scientists create first AI-designed viruses to fight drug-resistant superbugs
Euronews
The Guardian
ScienceAug 7, 2026 · 1 min
First Thing: Soaring oil profits from US-Iran conflict drive calls for windfall tax
The Guardian
Times of India
ScienceAug 7, 2026 · 1 min
Scientists discover Titan's methane rainstorms are more intense than expected
Times of India
The Guardian
ScienceAug 7, 2026 · 1 min
‘Walking sausage’: the giant stick insect that proves it’s not over till we give up
The Guardian