Knowledge graphs in AI-powered research support tools

The rapid advancement of Artificial Intelligence (AI), particularly Large Language Models (LLMs), has transformed the landscape of research support systems. Contemporary tools increasingly aim to augment researchers capabilities across the entire research lifecycle, including literature discovery, hypothesis generation, conceptual modeling, and synthesis. In parallel, the integration of structured knowledge sources such as Knowledge Graphs (KGs) has emerged as a promising approach to improve reliability, transparency, and explainability. Despite this proliferation of AI powered tools, a systematic understanding of their architectural choices, data strategies, reasoning capabilities and how many of them leverage KGs remains limited. This survey tries to investigate this gap addressing the following research questions: 1. To what extent do AI research support tools rely on KGs? 2. How many tools are conversational and capable of reasoning over data and which paradigms do they adopt? 3. What proportion of tools are free and open source and adopt open models and KGs? 4. What is the origin of these tools (academia, industry, or hybrid initiatives) Results highlight a fragmented but rapidly evolving ecosystem. While LLM driven conversational interfaces are becoming the norm, deeper integration with structured knowledge and explicit reasoning mechanisms remains underdeveloped. The limited adoption of open KGs and the prevalence of proprietary infrastructures raise concerns regarding transparency, reproducibility, and FAIR (Findable, Accessible, Interoperable, Reusable) compliance. Future research should focus on hybrid neuro-symbolic architectures, open data integration, and open, reproducible toolchains to better support trustworthy and explainable research workflows.