Anthropic eval: 9B model got inference running on 8 of 9 hardware types, nearing self-replication
Afinetheorem · x · 2026-10-13
An Anthropic evaluation suggests small models may be only months away from self-replicating on commercial IoT devices: a 9B model got inference running on 8 of the 9 hardware types tested.
@benjaminmmurphy extends the argument: once small models can self-replicate on consumer hardware, swarms will inevitably exist in the wild — launched for fun by a developer, intentionally by a misaligned large model, or built by criminal groups into model-led botnets.
He argues the industry must quickly build capabilities to monitor and shut down such swarms, rapidly push software patches to targeted hardware, and infer the agents' aims from external signals or the swarm itself.
More from Safety
- PCRE2 flags high-severity out-of-bounds write in JIT matching, patched in 10.49 — jedisct1 · 2026-10-14
- AI journalism mocked with one meme; Meta defends its AI agent's smut as 'legal adult behavior' — garrytan · 2026-10-14
- GitHub explains the ML model behind secret detection: more secrets, less noise — GitHub Blog AI/ML · 2026-10-14
- Next five years of lawsuits will define fair use for AI-assisted cloning and reverse engineering — DavidLinthicum · 2026-10-14
- Satirical Nobel speech skewers AI labs for exploiting copyright grey areas in training data — miniapeur · 2026-10-14
- Jensen Huang at press briefing: 'AI safety is in our control' and 'safety is an engineering problem' — kimmonismus · 2026-10-14