The 2025 AI Agent Index: Documenting Technical and Safety Features of Deployed Agentic AI Systems
Индекс ИИ-агентов 2025 года: документирование технических характеристик и характеристик безопасности развернутых агентных систем искусственного интеллекта
2026-06-23
SCID: 54.1/f3parzwy
Discuss with AI
AI Agent IndexAI agent capabilitiesAI safety featuresagentic AI systemsdeveloper transparency
Figures from the paper
Abstract (AI)
Agentic AI systems are increasingly capable of performing professional and personal tasks with limited human involvement. However, tracking these developments is difficult because the AI agent ecosystem is complex, rapidly evolving, and inconsistently documented, posing obstacles to both researchers and policymakers. To address these challenges, this paper presents the 2025 AI Agent Index. The Index documents information regarding the origins, design, capabilities, ecosystem, and safety features of 30 state-of-the-art AI agents based on publicly available information and email correspondence with developers. In addition to documenting information about individual agents, the Index illuminates broader trends in the development of agents, their capabilities, and the level of transparency of developers. Notably, we find that transparency varies substantially across agent developers and observe that most developers share little information about safety, evaluations, and societal impacts. The 2025 AI Agent Index is available online at https://aiagentindex.mit.edu.
Key Findings
1
Most developers disclose limited information about agent safety, evaluations, and societal impacts.
2
The 2025 AI Agent Index documents the origins, design, capabilities, ecosystems, and safety features of 30 state-of-the-art AI agents.
3
The Index combines publicly available information with email correspondence from developers to characterize deployed agentic AI systems.
4
The Index provides a resource for researchers and policymakers tracking a complex, rapidly evolving, and inconsistently documented AI agent ecosystem.
5
Transparency varies substantially among AI agent developers, particularly regarding technical and safety-related information.
Research Object
30 state-of-the-art deployed agentic AI systems
Research Subject
their origins, design, capabilities, ecosystem, safety features, and developer transparency
Publication Details
Publication Date
2026-06-23
Journal
Publisher
ISSN
Open access PDF
Access Type
Author Information
Download PDF
Subscribe to digest