Unverified60% confidenceOpinionExact time
Some researchers estimate that the knowledge half-life of applied AI technology is approximately 1 to 3 years, while the half-life of foundational mathematics and statistical theory is as long as several decades.
2
Sources
60%
Confidence
Long-term
Relevance
7/20/2026
First Seen
Sources
Related Claims
UnverifiedThe current capability iteration cycle for large AI models has shortened to 3-6 months, compared to Moore's Law's roughly 18-month cycle in traditional hardware72% similarUnverifiedThe recruitment requirement of '4 years of AI experience' is untenable in terms of timeline, because the technology for generative AI chatbot agents truly entered the mainstream only two to three years ago.70% similarUnverifiedThe AI evaluation trend of the past two years has gradually shifted from knowledge question answering to code races, Agent tasks, web operations, and then household chore simulation.69% similarUnverifiedThe trajectory of generative AI has broken the linear pattern in which most technologies take years or even decades to mature.68% similarVerifiedAI development is divided into three stages: 2012–2020 was perceptual intelligence, 2020 to the present is cognitive intelligence, and the next stage is action intelligence.68% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/566489API
curl https://kongchang.com/api/v1/knowledge/claims/566489MCP
get_claim(id=566489)