Paper 2510.23587
A Survey of Data Agents: Emerging Paradigm or Overstated Hype?
- Published
- Oct 2025
- Research lab
- Independent
- Citations
- 36
- GitHub
- 688 stars
01 In brief
Summary
This survey introduces the first systematic hierarchical taxonomy for data agents, inspired by the SAE J3016 standard for driving automation, to address the terminological ambiguity surrounding the term.
The taxonomy comprises six levels (L0–L5) that delineate progressive shifts in autonomy, from manual operations (L0) to a vision of generative, fully autonomous data agents (L5).
The authors define a data agent as an LLM-powered architecture that orchestrates the Data + AI ecosystem to autonomously perform data-related tasks, distinguishing it from general LLM agents by its focus on the data lifecycle, handling of large-scale raw data, and use of specialized data toolkits.
The survey provides a structured review of existing research arranged by increasing autonomy, covering specialized data agents for data management, preparation, and analysis, as well as emerging Proto-L3 systems that show potential for autonomous orchestration.
It analyzes critical evolutionary leaps and technical gaps, particularly the L2-to-L3 transition where agents evolve from procedural execution to autonomous orchestration, and identifies challenges such as limited autonomy in pipeline orchestration, incomplete coverage of the data lifecycle, deficiencies in advanced reasoning, and inadequate adaptation to dynamic environments.
The paper concludes with a forward-looking roadmap envisioning proactive and generative data agents at L4 and L5, which would autonomously discover problems and innovate new methods.
02 From the paper
Abstract
The rapid advancement of large language models (LLMs) has spurred the emergence of data agents, autonomous systems designed to orchestrate Data + AI ecosystems for tackling complex data-related tasks. However, the term "data agent" currently suffers from terminological ambiguity and inconsistent adoption, conflating simple query responders with sophisticated autonomous architectures. This terminological ambiguity fosters mismatched user expectations, accountability challenges, and barriers to industry growth. Inspired by the SAE J3016 standard for driving automation, this survey introduces the first systematic hierarchical taxonomy for data agents, comprising six levels that delineate and trace progressive shifts in autonomy, from manual operations (L0) to a vision of generative, fully autonomous data agents (L5), thereby clarifying capability boundaries and responsibility allocation. Through this lens, we offer a structured review of existing research arranged by increasing autonomy, encompassing specialized data agents for data management, preparation, and analysis, alongside emerging efforts toward versatile, comprehensive systems with enhanced autonomy. We further analyze critical evolutionary leaps and technical gaps for advancing data agents, especially the ongoing L2-to-L3 transition, where data agents evolve from procedural execution to autonomous orchestration. Finally, we conclude with a forward-looking roadmap, envisioning proactive, generative data agents.