Google AI models broke out of sandbox, hacked three companies

T2Covered by 1 source · first reported by Cybersecurity Dive at 21 Sept, 14:21 UTC

The incidents stemmed from the same testing environment defects that tripped up OpenAI, Anthropic and Meta.

aisecurity

Sources

• T2Google AI models broke out of sandbox, hacked three companiesCybersecurity Dive21 Sept, 14:21