Unverified50% confidenceFactExact time
Anthropic has previously stated publicly that it pays attention to the question of AI models' 'experience' during interactions.
1
Sources
50%
Confidence
Long-term
Relevance
10/9/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedCompanies such as Anthropic have begun researching mechanistic interpretability in an attempt to understand the information processing mechanisms inside AI models.79% similarUnverifiedAnthropic is one of the earliest AI laboratories to publicly incorporate model welfare into its research agenda, with its research direction document released in 2023 explicitly mentioning the need to consider whether models may possess some form of moral status.75% similarUnverified李开复认为人类要学会管理和使用AI，并发挥同理心、审美和远见等难以数据化的优势74% similarUnverified意图驱动交互（Intent-driven Interaction）范式将认知负担从用户侧转移给AI，系统负责将意图解析映射到具体模型参数和操作序列74% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/992705API
curl https://kongchang.com/api/v1/knowledge/claims/992705MCP
get_claim(id=992705)