Vibe Researching as Wolf Coming: Can AI Agents with Skills Replace or Augment Social Scientists?
#AI agents #social science #vibe researching #research automation #cognitive task framework #scholar-skill #artificial intelligence #research methodology
📌 Key Takeaways
- AI agents can now execute entire research pipelines autonomously
- Vibe researching is introduced as an AI-era parallel to vibe coding
- AI excels at speed and methodology but lacks theoretical originality
- Three implications include fragile augmentation, stratification risk, and pedagogical crisis
📖 Full Retelling
Yongjun Zhang published a groundbreaking research paper titled 'Vibe Researching as Wolf Coming: Can AI Agents with Skills Replace or Augment Social Scientists?' on arXiv on February 25, 2026, examining how advanced AI systems might transform social science research methodologies. The paper introduces the concept of 'vibe researching' as an AI-era parallel to 'vibe coding,' utilizing a 21-skill plugin called 'scholar-skill' that covers the entire research pipeline from idea to submission. Unlike previous chatbot technologies that responded to isolated queries, these new AI agents can autonomously execute multi-step reasoning workflows with persistent state, tool access, and specialist skills, including reading files, running code, querying databases, and searching the web.
Zhang develops a cognitive task framework that classifies research activities along two dimensions: codifiability and tacit knowledge requirement. This framework identifies a delegation boundary that is cognitive rather than sequential, meaning it cuts through every stage of the research pipeline rather than occurring between stages. The analysis reveals that AI agents excel at speed, coverage, and methodological scaffolding but struggle with theoretical originality and tacit field knowledge that human researchers possess through experience and domain expertise.
The paper concludes with three significant implications for the social science profession: augmentation with fragile conditions where human oversight remains essential, risk of stratification where access to AI tools creates inequalities, and a potential pedagogical crisis as traditional training methods become obsolete. To address these challenges, Zhang proposes five principles for responsible 'vibe researching' that aim to balance technological advancement with maintaining the integrity and value of human insight in social science research.
🏷️ Themes
AI in Research, Social Science Transformation, Human-AI Collaboration
📚 Related People & Topics
AI agent
Systems that perform tasks without human intervention
In the context of generative artificial intelligence, AI agents (also referred to as compound AI systems or agentic AI) are a class of intelligent agents distinguished by their ability to operate autonomously in complex environments. Agentic AI tools prioritize decision-making over content creation ...
Entity Intersection Graph
Connections for AI agent:
🏢
OpenAI
5 shared
🌐
Large language model
4 shared
🌐
OpenClaw
2 shared
🌐
Artificial intelligence
2 shared
🌐
Workflow
1 shared
Original Source
--> Computer Science > Artificial Intelligence arXiv:2602.22401 [Submitted on 25 Feb 2026] Title: Vibe Researching as Wolf Coming: Can AI Agents with Skills Replace or Augment Social Zhang View a PDF of the paper titled Vibe Researching as Wolf Coming: Can AI Agents with Skills Replace or Augment Social Scientists?, by Yongjun Zhang View PDF HTML Abstract: AI agents -- systems that execute multi-step reasoning workflows with persistent state, tool access, and specialist skills -- represent a qualitative shift from prior automation technologies in social science. Unlike chatbots that respond to isolated queries, AI agents can now read files, run code, query databases, search the web, and invoke domain-specific skills to execute entire research pipelines autonomously. This paper introduces the concept of vibe researching -- the AI-era parallel to ``vibe coding'' (Karpathy, 2025) -- and uses scholar-skill, a 21-skill plugin for Claude Code covering the full research pipeline from idea to submission, as an illustrative case. I develop a cognitive task framework that classifies research activities along two dimensions -- codifiability and tacit knowledge requirement -- to identify a delegation boundary that is cognitive, not sequential: it cuts through every stage of the research pipeline, not between stages. I argue that AI agents excel at speed, coverage, and methodological scaffolding but struggle with theoretical originality and tacit field knowledge. The paper concludes with an analysis of three implications for the profession -- augmentation with fragile conditions, stratification risk, and a pedagogical crisis -- and proposes five principles for responsible vibe researching. Comments: Commentary Subjects: Artificial Intelligence (cs.AI) ; Human-Computer Interaction (cs.HC) Cite as: arXiv:2602.22401 [cs.AI] (or arXiv:2602.22401v1 [cs.AI] for this version) https://doi.org/10.48550/arXiv.2602.22401 Focus to learn more arXiv-issued DOI via DataCite (pending registrati...
Read full article at source