Unverified50% confidenceOpinionExact time
The creator questions whether strong artificial intelligence must necessarily be realized in the form of LLMs, holding that the LLM path is currently going smoothly but may not be the only path.
1
Sources
50%
Confidence
Long-term
Relevance
11/4/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedAI cannot be the sole dependency in a critical path; the fix philosophy is rules engine first, with LLM only as a fallback.73% similarUnverifiedYou don't need to first become an excellent ML engineer to enter the Agentic AI direction, but you do need a pragmatic understanding of the boundaries of LLM capabilities.71% similarUnverifiedCurrent LLM agents are good at execution but not good at judging whether they are still on track, especially in long-horizon tasks and multi-agent collaboration.71% similarUnverifiedBefore building an agent, a reliable direct LLM call baseline should be established, and autonomy should be introduced only when the remaining failures are about "choosing a path."69% similarUnverifiedThe security risk of AI agent architecture lies in the fact that LLM decision-making is probabilistic rather than deterministic, lacks an independent verification layer, and the LLM acts as both policy maker and execution initiator, violating the principle of separation of duties.69% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1134879API
curl https://kongchang.com/api/v1/knowledge/claims/1134879MCP
get_claim(id=1134879)