Giving AI Agents Access to Cryptocurrency and Smart Contracts Creates New Vectors of AI Harm
Not on the map. Why?
Abstract (arXiv)
There is growing interest in giving AI agents access to cryptocurrencies and smart contracts. But doing so, this position paper argues, could spawn categorically new vectors of AI harm. To support this notion, we first examine the unique technical properties of cryptocurrencies and smart contracts that, when paired with AI agents, make this so. Next, we describe three of these new vectors of AI harm in detail, providing a first-of-its-kind taxonomy. Finally, we call for more research into ways to prevent or mitigate these risks, proposing several as a starting point.
Key facts. Cited by 3 papers in the map (data through 7 Oct 2026). Builds on 37 of its 224 references that are in the map. First posted on arXiv 11 Jul 2025.
Builds on
37 of 224 references.
References · 37
- 2023Toolformer: Language Models Can Teach Themselves to Use Toolscited by 4,072
- 2016Concrete Problems in AI Safetycited by 1,582
- 2023AgentBench: Evaluating LLMs as Agentscited by 1,211
- 2021Ethical and social risks of harm from Language Modelscited by 957
- 2018Scalable agent alignment via reward modeling: a research directioncited by 438
- 2023Large Language Models as Tool Makerscited by 293
- 2022The Alignment Problem from a Deep Learning Perspectivecited by 246
- 2023Large Language Model Alignment: A Surveycited by 241
- 2024Agent AI: Surveying the Horizons of Multimodal Interactioncited by 167
- 2018The Malicious Use of Artificial Intelligence: Forecasting, Prevention, and Mitigationcited by 159
- 2025Emergent Misalignment: Narrow finetuning can produce broadly misaligned LLMscited by 158
- 2023Harms from Increasingly Agentic Algorithmic Systemscited by 134
25 more
- 2023Are Emergent Abilities in Large Language Models just In-Context Learning?cited by 109
- 2023Evaluating Language-Model Agents on Realistic Autonomous Taskscited by 94
- 2024Feedback Loops With Language Models Drive In-Context Reward Hackingcited by 70
- 2025LLM Agents Making Agent Toolscited by 45
- 2024Breaking Down the Defenses: A Comparative Survey of Attacks on Large Language Modelscited by 42
- 2024The Reasons that Agents Act: Intention and Instrumental Goalscited by 23
- 2024Tools Fail: Detecting Silent Errors in Faulty Toolscited by 23
- 2025Real AI Agents with Fake Memories: Fatal Context Manipulation Attacks on Web3 Agentscited by 23
- 2023Position: Key Claims in LLM Research Have a Long Tail of Footnotescited by 20
- 2025LLM-Powered Multi-Agent System for Automated Crypto Portfolio Managementcited by 20
- 2017Dissecting Ponzi schemes on Ethereum: identification, analysis, and impactcited by 19
- 2020Charting the Landscape of Online Cryptocurrency Manipulationcited by 15
- 2025Eliza: A Web3 friendly AI Agent Operating Systemcited by 14
- 2019The Art of The Scam: Demystifying Honeypots in Ethereum Smart Contractscited by 13
- 2024Security Threats in Agentic AI Systemcited by 11
- 2024A Reflective LLM-based Agent to Guide Zero-shot Cryptocurrency Tradingcited by 10
- 2023Increased Compute Efficiency and the Diffusion of AI Capabilitiescited by 8
- 2021Forsage: Anatomy of a Smart-Contract Pyramid Schemecited by 4
- 2023SoK: Design, Vulnerabilities, and Security Measures of Cryptocurrency Walletscited by 4
- 2024Efficacy of Various Large Language Models in Generating Smart Contractscited by 4
- 2018The ICO Phenomenon and Its Relationships with Ethereum Smart Contract Environmentcited by 2
- 2023Does Money Laundering on Ethereum Have Traditional Traits?cited by 2
- 2024The Feasibility of a Smart Contract "Kill Switch"cited by 2
- 2023Cross Chain Bribery Contracts: Majority vs Mighty Minoritycited by 1
- 2024Liquefaction: Privately Liquefying Blockchain Assetscited by 1
Built on by
Citing papers per year: 2026: 3.
3 papers.