Can an AI copy itself? Inside Anthropic’s 15-TV-box experiment
This story is only covered by news sources that have yet to be evaluated by the independent media monitoring agencies we use to assess the quality and reliability of news outlets on our platform. Learn more here.4 Articles
4 Articles
Anthropic ran a 26B model across 15 Android TV boxes, and got 0.39 tokens a second
Anthropic's Frontier Red Team ran a 26-billion-parameter model across 15 off-brand Android TV boxes, bought for about $55 each with 4GB of RAM apiece, and got 0.39 tokens a second out of the lot. William Zhou and Daniel Freeman write that the model was Gemma4-26B-A4B, that the result shows "signs of life for AI worms", ... Anthropic ran a 26B model across 15 Android TV boxes, and got 0.39 tokens a second The post Anthropic ran a 26B model across…
This is happening in the AI market. Even cheap TV boxes can run certain models, and we already know that there are agent-based mechanisms that can organize self-copying between machines. Anthropic's experiments point to a rather dangerous scenario, but they give us (for now) hope: for it to come true, we need much more powerful hardware.
Can an AI copy itself? Inside Anthropic’s 15-TV-box experiment
Ashu Kumar, writing at The Agentic Stack, covers Anthropic Frontier Red Team research published October 12 on whether an AI system can replicate itself onto cheap hardware. The team split Gemma4-26B-A4B across 15 off-brand Android TV boxes, each about $55 with 4 GB of memory, and got 0.39 generated tokens per second. Slow for a […]
Anthropic Ran an AI Across 15 Smart TV Boxes to See If It Could Spread
Anthropic’s Frontier Red Team bought 15 off-brand Android TV boxes, roughly $55 each with 4GB of RAM, wired them together, and split a 26-billion-parameter model across them. Each box held a couple of the model’s layers. The fleet produced 0.39 tokens per second. That number anchors the research published on 12 October by William Zhou […] The post Anthropic Ran an AI Across 15 Smart TV Boxes to See If It Could Spread appeared first on H2S Media.
Coverage Details
Bias Distribution
- There is no tracked Bias information for the sources covering this story.
Factuality
To view factuality data please Upgrade to Premium
