狼来了:具备技能的人工智能代理能否取代或增强社会科学家?
📄 中文摘要
AI代理是执行多步骤推理工作流程的系统,具备持久状态、工具访问和专业技能,标志着社会科学领域自动化技术的质变。与仅能响应孤立查询的聊天机器人不同,AI代理能够自主读取文件、运行代码、查询数据库、搜索网络,并调用领域特定技能,执行完整的研究流程。研究提出了“氛围研究”的概念,作为AI时代的“氛围编码”平行概念,并以scholar-skill为例,展示了涵盖从创意到提交的完整研究流程的21项技能插件。构建了一个认知任务框架,对研究活动进行分类,为社会科学的未来研究提供了新的视角。
📄 English Summary
Vibe Researching as Wolf Coming: Can AI Agents with Skills Replace or Augment Social Scientists?
AI agents represent a qualitative shift in social science automation, executing multi-step reasoning workflows with persistent state, tool access, and specialized skills. Unlike chatbots that handle isolated queries, AI agents can autonomously read files, run code, query databases, search the web, and invoke domain-specific skills to execute entire research pipelines. The concept of 'vibe researching' is introduced as a parallel to 'vibe coding' in the AI era, illustrated by the scholar-skill plugin, which encompasses 21 skills covering the full research pipeline from idea generation to submission. A cognitive task framework is developed to classify research activities, providing a new perspective on the future of research in social sciences.
Powered by Cloudflare Workers + Payload CMS + Claude 3.5
数据源: OpenAI, Google AI, DeepMind, AWS ML Blog, HuggingFace 等