AI CRYPTO

Sygnum Becomes First Swiss Bank to Execute Live On-Chain Transactions via AI Agent

Sygnum has pioneered the use of AI agents for live on-chain transactions, enabling clients to retain custody while enhancing transaction efficiency.

Sygnum Becomes First Swiss Bank to Execute Live On-Chain Transactions via AI Agent
CoinSynaptic Desk
AI CRYPTO · Correspondent
· PUBLISHED MAY 18, 2026 · UPDATED 12:06 ET · 2 MIN READ

Sygnum Bank has achieved a milestone by becoming the first regulated bank in Switzerland to use an AI agent for executing live on-chain transactions. This innovation enables clients to retain control over their assets while utilizing artificial intelligence to improve transaction processes.

The AI agent operates through Sygnum's proprietary Model Context Protocol (MCP) server, powered by Anthropic’s Claude AI model. Clients can request transactions using everyday language, allowing the agent to plan necessary steps, review smart contracts, and identify potential risks before presenting the transaction for client approval. Importantly, all signing processes occur on the client’s device, ensuring that private keys remain secure and never leave their possession.

The Shift Towards Transactional AI Agents

Sygnum’s initiative reflects a broader trend in the banking sector, where institutions are increasingly moving AI agents beyond advisory roles into active transaction execution. This shift is evident among traditional banks and digital asset custodians as financial institutions rush to adopt AI technologies that streamline operations and enhance client experiences.

https://www.youtube.com/watch?v=GbrXDPZaPdU

Thomas Frei, Head of AI and Data Analytics at Sygnum, underscored the importance of connecting AI agents to wallets, stating, "Connecting AI agents to wallets is foundational to where finance is heading. The next decade will see agents transacting, settling and interacting with markets on behalf of clients." This vision points to a future where AI plays a critical role in financial decision-making and execution.

https://www.youtube.com/watch?v=XuxbaLx-7BM

Illustrative visual for: Sygnum Becomes First Swiss Bank to Execute Live On-Chain Transactions via AI Age

Use Cases and Implications

https://www.youtube.com/watch?v=y96zV3vHZNo

The practical applications for Sygnum's AI agent are varied and impactful. Clients can engage in activities such as stablecoin transactions, asset swaps, on-chain lending, token wrapping, and liquidity provisioning. As these use cases expand, the role of AI in financial transactions is likely to increase, emphasizing both efficiency and security.

See also  AI Agents Transform Security Operations with MCP Protocol

https://www.youtube.com/watch?v=TVqnZzkdibo

Sygnum is not alone in this endeavor. Other players, including Anchorage Digital, have begun exploring similar functionalities. In May, Anchorage launched its Agentic Banking platform, designed to facilitate fund transfers through regulated banking channels. Additionally, partnerships between FIS and Anthropic focus on developing agentic AI for banking applications, including anti-money-laundering efforts.

Looking Ahead

https://www.youtube.com/watch?v=SgaoBrPHdYk

The deployment of AI agents for live transactions marks a significant step for the banking industry. As more institutions adopt these technologies, the potential for AI to transform financial services will continue to grow. By integrating AI into the transaction layer, banks are enhancing operational efficiency and paving the way for a new era of automated financial interactions that prioritize client control and security.

https://www.youtube.com/watch?v=56gEUZiOSmE

CoinSynaptic Desk

AI Crypto · 2,198 stories

CoinSynaptic Desk covers the intersection of artificial intelligence and decentralized networks — frontier AI infrastructure, crypto-native AI agents, Bittensor subnets, DePIN economies, and tokenized compute.

THE DAILY SIGNAL

The stories that move AI & crypto markets — before the market reacts.

Free. 7am ET. Five stories. 62,400 readers.