Unverified50% confidenceFactExact time
In Anthropic's experiments, models with fewer than 30 billion parameters were already able to complete most of the individual steps required for AI worm replication, but still made errors on the complete end-to-end chain.
1
Sources
50%
Confidence
Long-term
Relevance
10/17/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedAnthropic predicts that if small models maintain their current rate of improvement, they will possess complete end-to-end AI worm replication capability before the end of the year.78% similarUnverifiedMathematical models of self-replicating AI worms show that if each agent can infect multiple new devices, the propagation chain may grow exponentially, making it difficult for defenders to block it through single-point removal.74% similarUnverifiedA clear trend toward smaller models has emerged in the AI field, ranging from the Phi series to various high-efficiency models with 1–3B parameters.69% similarUnverified普林斯顿研究让AI代理模拟创业500天，结果绝大多数破产仅3个存活，简单规则算法反而战胜了所有大模型69% similarUnverifiedAI agents may need to execute dozens or even hundreds of consecutive steps, and any network interruption or crash could cause the entire task to fail.68% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1054670API
curl https://kongchang.com/api/v1/knowledge/claims/1054670MCP
get_claim(id=1054670)