Financial technology is moving toward a new architecture. Artificial intelligence is automating analysis and financial operations, while stablecoins and tokenization are changing how money and assets can move across digital networks. The Financial Stability Board reported in June 2026 that financial institutions are already using AI to transform operations and services, while warning that rapid adoption can introduce or amplify financial risks.
Regulation is starting to adjust to that shift. On August 18, 2026, the U.S. Securities and Exchange Commission proposed Regulation Crypto Assets. The framework would create tailored offering rules for certain investment contracts involving crypto assets, including two registration exemptions and a conditional safe harbor. It signals that regulators are considering rules designed around digital assets rather than simply applying older structures unchanged.
What Does the New Financial Stack Look Like?
The emerging system can be understood as several connected layers. AI interfaces and software agents sit near the top, interpreting instructions and making decisions. Below them, digital identity and compliance systems determine who can transact. Updated guidance from the National Institute of Standards and Technology covers identity proofing, authentication and federation, highlighting the importance of reliable identity controls in digital systems.
Payment rails, stablecoins and tokenized deposits can then provide different ways to transfer value. Tokenized assets add securities, bonds or other financial claims that exist as programmable digital representations. Blockchains or similar ledgers provide transaction infrastructure, while banks, custodians and central banks can remain connected underneath.
The Bank for International Settlements argues that tokenization could improve cross-border payments and securities markets. Its proposed next-generation model combines tokenized central bank reserves, commercial bank money and government bonds. However, the BIS also warns that current stablecoins can suffer from fragmentation and other weaknesses that limit their ability to function as the foundation of the monetary system.
What Changes When Software Can Move Money?
The convergence becomes more significant when AI moves from giving advice to taking permitted actions. A company could eventually authorize an AI agent to monitor cash positions, calculate upcoming obligations and initiate approved transfers. Another agent might compare tokenized investments, execute transactions within defined limits and produce compliance records.
This could make treasury operations and financial markets more responsive. Stablecoins may provide programmable payment instruments, while tokenized securities could make assets easier for software systems to interact with. The International Monetary Fund notes that stablecoins may improve some payments, especially across borders, although their global reach also creates challenges involving capital flows, oversight and financial stability.
Why Human Control Still Matters
More automation also creates a larger security problem. An incorrect AI decision, compromised identity credential or flawed smart contract could trigger transactions faster than people can intervene. Financial institutions therefore need controls covering authorization, cybersecurity, model oversight, transaction limits and accountability. The Financial Stability Board has emphasized governance and safeguards throughout the AI lifecycle as financial firms increase adoption.
The future financial stack may therefore be highly automated without becoming fully autonomous. AI could decide what action to recommend, digital identity could establish permission, stablecoins could provide settlement and blockchains could record ownership. Yet banks, regulators and human operators may still define the boundaries. The central challenge is making programmable finance faster without allowing speed and autonomy to outrun security, accountability and trust.