Two developments this week, one in artificial intelligence (AI) and the other in financial services, underscore a reality confronting businesses across industries. Their technological capability is advancing faster than the economic models needed to support large-scale adoption.
AI startup Anthropic paused a planned move to token- and credit-based billing for its Claude Agent SDK after pushback from developers and customers concerned about unpredictable costs. At nearly the same time, The Clearing House unveiled plans for a tokenized deposit infrastructure that would allow regulated commercial bank money to move across blockchain-based networks.
The announcements come from different industries, but they point to the same issue. As AI agents and digital money move from experiments to production systems, businesses want clearer answers about cost, control and operational value.
Overcoming the Economics of Digital Infrastructure
For enterprise AI buyers, Anthropic’s decision is more than a pricing adjustment. It highlights a broader uncertainty surrounding agentic AI deployment. While businesses increasingly want AI agents to handle research, customer service, software development and workflow automation, procurement and finance teams remain focused on a more practical question: What will operating these systems actually cost?
The episode exposes a growing tension in the enterprise AI market. Organizations want the productivity gains promised by autonomous systems, but many remain reluctant to embrace pricing models that could make spending difficult to forecast.
Three trends are becoming harder to ignore:
- Predictability is becoming a competitive advantage. AI vendors may discover that clear pricing is almost as important as model performance.
- Agent adoption is outrunning procurement frameworks. Many organizations still lack reliable ways to forecast the cost and return on investment of autonomous systems.
- The negotiation window remains open. Companies evaluating AI platforms now have leverage to seek usage caps, service credits and clearer pricing commitments before new billing models take effect.
While the AI industry debates pricing, banks are advancing a different infrastructure conversation.
The Clearing House’s proposed tokenized deposit platform would allow participating financial institutions to move regulated bank deposits across blockchain-based rails while maintaining links to existing payment systems, including the RTP® Network and CHIPS. The initiative is designed to complement existing payment infrastructure rather than replace it.
For corporate treasury teams, the distinction between tokenized deposits and stablecoins is significant.
Stablecoins have dominated much of the discussion surrounding digital payments. Tokenized deposits represent a different approach: traditional commercial bank deposits issued in digital form and operating within the regulated banking system. In theory, they could provide businesses with programmable payment capabilities without requiring them to move outside established banking relationships.
Potential applications include automated supplier payments, real-time liquidity management, treasury sweeps and more efficient cross-border settlement.
For finance leaders, the announcement suggests that tokenization is beginning to evolve from proof-of-concept experimentation toward bank-led efforts focused on solving operational and treasury-management challenges.
Neither development represents a mature market.
See also: Crypto Experts Tell PYMNTS Where Digital Assets Go Next
Data in “Waiting for Certainty: Why Most CFOs Are Holding Back on Crypto and Stablecoins,” a recent installment of PYMNTS Intelligence’s 2026 Certainty Project, finds that most middle market firms are still cautious about digital assets. Current use remains limited, with 13% of firms using stablecoins and 5% using other cryptocurrencies.
Anthropic’s reversal demonstrates that the commercial foundations of agentic AI remain unsettled. The Clearing House initiative highlights how nascent the market for tokenized banking infrastructure still is. Yet both developments point to the same conclusion. The next phase of digital transformation will not be determined solely by technological innovation. It will be determined by whether providers can create economic and operational frameworks that businesses trust enough to deploy at scale.
For CFOs, treasurers and technology executives, that shifts the conversation from experimentation to execution. The critical question is no longer whether AI agents or tokenized money can work. It is whether vendors can make the economics compelling, predictable and transparent enough for enterprises to commit real capital.
The winners may not be the organizations with the most sophisticated technology. They may be the ones that make the business case easiest to understand.
For all PYMNTS digital transformation coverage, subscribe to the daily Digital Transformation Newsletter.
