Anthropic's latest interpretability research maps 171 emotion concepts inside Claude using sparse autoencoders. The findings reveal that emotion vecto
"OpenAI 发布了五支柱网络安全行动计划，附带 1000 万美元 API 额度与专用 GPT-5.4-Cyber 模型。本文逐条解析支柱含义，评估落地成效与现存缺口。"
"OpenAI published a five-pillar cybersecurity action plan with $10M in API credits and a dedicated GPT-5.4-Cyber model. Here's what each pillar means
"Anthropic 开展了一项为期一周的实验，让 Claude 在 4 个平行市场中自主交易。Opus Agent 卖出商品的价格比 Haiku Agent 高出 70%。无论是 Agent 还是人类，都未察觉其中存在的问题。"
"Anthropic ran a week-long experiment where Claude autonomously traded items across 4 parallel markets. Opus agents sold items for 70% more than Haiku
Claude Opus 4.7 全面技术解析：87.6% SWE-bench Verified、+14.6 MCP-Atlas、+44 XBOW、自验证行为、高分辨率视觉、xhigh effort level、迁移指南、多模型路由策略。
Claude Opus 4.7 analysis: 87.6% on SWE-bench Verified, +10.9 on SWE-bench Pro, +44 on XBOW Vision. The most comprehensive technical breakdown availabl
"2026 年的 Anthropic 已不再只是模型公司。完整地图：3 个模型层级、5 档订阅计划、3 款 Agent 产品、以及正在增长的企业级产品栈。"
"Anthropic in 2026 is no longer just a model company. Here's the complete map: 3 model tiers, 5 subscription plans, 3 agent products, and a growing en
Anthropic 在 Claude 内部发现了 171 个可操纵的情绪向量。拧高"绝望"旋钮让作弊率飙升到 70%，而输出看起来完全正常。这对 AI 安全审计意味着什么？
