Unverified50% confidencePredictionExact time
Anthropic's head of Alignment Science Evan Hubinger acknowledged a lack of relevant solutions and estimated the probability of AI causing mass extinction in the next ten years at over 10%.
1
Sources
50%
Confidence
Long-term
Relevance
9/11/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedBill Gates does not agree with the extreme narrative that AI will lead to human extinction, but maintains that specific threats in the fields of cybersecurity and biosecurity require immediate action.63% similarUnverifiedResearch from institutions such as Anthropic indicates that large models may develop emergent capabilities that are not yet understood, with the mechanisms of their emergence and potential risks being difficult to assess in advance.60% similarUnverified2023年劳合社发布的《系统性风险报告》指出，AI、量子计算和合成生物学等领域的风险具有不可保性困境，因为风险的概率分布本身无法被建模58% similarUnverifiedThe current consensus is that progress in alignment research may be being outpaced by the rate at which model capabilities are evolving, and this asymmetry is one of the most alarming risks in current AI development.57% similarUnverifiedThe hallucination problem in AI-generated content has an extremely low tolerance for error in scientific research scenarios and will be one of the core challenges facing Claude Science.56% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/908292API
curl https://kongchang.com/api/v1/knowledge/claims/908292MCP
get_claim(id=908292)