Anthropic's Frontier Red Team reports that small AI models could soon self-replicate on commercial IoT devices like smart ovens and off-brand Android TV boxes.
1 editorial report · 1 verified social mention. The most authoritative report leads while later evidence completes the story.
Anthropic warned that smaller AI models could gain the capability to self-replicate across connected devices within months. Company experiments demonstrated that malicious AI agents can combine computing power from affordable devices to run additional models, with models under 30 billion parameters already capable of executing many replication tasks. Simplified tests without certain safety measures allowed model… https:// en.hacks.gr/proeidopoiisi-anth ropic-ta-mikrotera-systimata-technitis-noimosynis-mporei-na-aytoanaparagontai-se-oikiakes-syskeyes/ # Anthropic # ArtificialIntelligence # MachineLearning # AI
Open mention