Bank of America AI and Digital Assets Expansion: What It Means for Financial Innovation

Bank of America AI and digital assets expansion is not a side project. It is a bank-scale rebuild of client service, treasury workflows, wealth management, market infrastructure, and settlement operations. Reuters has reported that the bank plans to spend billions of dollars on technologies such as AI, and Bloomberg has noted new senior roles focused on crypto, digital assets, and AI across global markets.

The interesting part is not that a large bank is testing AI. Everyone is. The signal is scale. Bank of America spends roughly $13 billion to $13.5 billion a year on technology, with about $4 billion directed to strategic initiatives that include AI. That budget makes its choices a useful indicator for where regulated finance is moving next.

Bank of America AI and Digital Assets Strategy at a Glance

The bank is expanding on two connected fronts. First, AI is being embedded into consumer banking, corporate payments, wealth management, investment banking, and markets workflows. Second, digital asset leadership is being formalized around tokenized deposits, stablecoins, digital collateral, custody, and trading settlement.

That combination matters. AI improves decisioning, search, service, fraud monitoring, and employee productivity. Tokenization changes the rails on which money, collateral, and securities can move. Put them together and you get the outline of a different financial stack: data-rich, automated, regulated, and much closer to real time.

AI Investment Is Moving From Pilots to Production

Bank of America has reportedly explored more than 45 generative AI proof-of-concept use cases, with 15 commercially live by late 2025. That ratio is worth a second look. In banking, most AI demos never survive model risk review, data controls, compliance checks, or integration with legacy systems.

Aditya Bhasin, Bank of America's Chief Technology and Information Officer, has said AI is having a transformative effect on employee efficiency and operational excellence. Strip away the executive phrasing and the point is simple. AI is being used where it cuts wait time, improves search, and helps staff answer complex questions faster.

A practical detail matters here. In enterprise banking AI, the model is rarely the hardest part. Retrieval quality is. If a payments assistant pulls from stale product documentation, even a top-tier large language model will produce a confident but wrong answer. In production retrieval-augmented generation systems, changing document chunk sizes from 300 tokens to 1,000 tokens can quietly change whether a term sheet exception is found or missed. Banks know this. That is why governance and source control matter as much as model selection.

Consumer Banking: Erica Shows What AI at Scale Looks Like

Bank of America's virtual assistant Erica has passed 3 billion client interactions since its 2018 launch. The bank reports more than 58 million Erica interactions per month and more than 20 million users.

Erica handles routine but high-volume tasks, including:

• Balance and transaction questions
• Spending pattern alerts
• Possible duplicate charge detection
• Basic budgeting guidance
• Routine mobile banking support

This is where AI pays for itself first. A single balance question is low value for a human agent, but millions of such questions create real cost. AI can absorb that load while human staff handle disputes, lending questions, fraud cases, and relationship work.

The bank has also reported about 30 billion digital client connections in the latest year through logins and proactive alerts, up 14 percent year over year. That number says something important. Digital banking is no longer just an app channel. It is becoming the primary client interface.

Corporate Banking: AskGPS and CashPro Bring AI to Treasury

Consumer chatbots get attention, but corporate banking may be the more serious test. Bank of America's Ask Global Payments Solutions, known as AskGPS, supports teams serving more than 40,000 business clients worldwide. It is trained on more than 3,200 internal documents and presentations, including product guides, term sheets, and FAQs.

For a corporate payments specialist, that is not a small convenience. Cross-border payments questions often depend on jurisdiction, cut-off time, currency, account structure, and client entitlement. A good AI assistant reduces the time spent digging through PDFs and internal portals.

CashPro is another key example. Bank of America has said about 65 percent of corporate clients using CashPro employ AI, representing around 40 percent of request volume. AI-driven CashPro Chat and transaction search have also seen record adoption.

For treasury teams, the impact is direct:

• Faster payment status checks
• Better search across transaction history
• Less manual exception handling
• Improved liquidity visibility
• Quicker answers to operational questions

To be blunt, this is where banks can win or lose corporate clients. A treasurer does not care that a platform uses generative AI. They care whether they can find a missing payment before the supplier calls again.

Wealth Management: AI Enters the Advisor Workflow

In March 2026, Merrill Wealth Management and Bank of America Private Bank rolled out AI-Powered Meeting Journey at full scale. The system helps advisors prepare for meetings, conduct client conversations, and manage follow-ups.

This is a different class of AI use case. It is not pure self-service. It supports a human advisor who needs context: portfolio changes, client life events, planning topics, risk flags, and prior meeting notes.

The trade-off is clear. AI can help advisors serve more clients with better preparation. But it should not replace fiduciary judgment. In wealth management, a technically correct recommendation can still be unsuitable if it ignores client goals, family constraints, tax exposure, or liquidity needs.

Digital Assets: Why the New Leadership Structure Matters

Bank of America has appointed Sonali Theisen as Global Head of Digital Asset Platforms. Her remit includes design, development, expansion, and governance of digital asset platforms, with integration into traditional financial infrastructure. She also continues to lead Global FICC electronic trading and markets strategic investments, which links digital assets directly to institutional markets.

Adam Dixon has been named Global Head of Digital Asset Transformation, with focus areas that include tokenized deposits, stablecoins, digital collateral circulation, encrypted trading settlement, and custody services. Kevin Milsom has been appointed Platforms AI Transformation Leader for global markets, driving AI adoption across business platforms and daily operations.

This is not just org-chart trivia. Large banks signal priorities through reporting lines, control ownership, and budget. Creating dedicated digital asset and AI transformation roles suggests the bank is preparing for production infrastructure, not isolated blockchain experiments.

Tokenized Deposits, Stablecoins, and Settlement

Bank of America has indicated that it is developing stablecoin solutions, while also noting that user acceptance remains a challenge. That caveat is realistic. Stablecoins are useful, but mainstream clients need trust, legal clarity, accounting treatment, redemption confidence, and simple user experience.

The bigger institutional story may be tokenized deposits and collateral. A tokenized deposit represents a bank deposit on blockchain-based rails. Digital collateral can move between counterparties, clearing venues, or custodians with better tracking and potentially faster settlement.

Possible benefits include:

• Shorter settlement cycles: Tokenized assets can reduce delays in post-trade processes if legal and operational frameworks are aligned.
• Programmable payments: Conditions can be attached to payment flows, useful for escrow, delivery-versus-payment, and automated treasury actions.
• Improved collateral mobility: Firms can move eligible collateral faster, which may reduce liquidity drag.
• Better auditability: Shared ledgers can improve visibility, although privacy controls are essential.

There is a warning here. Tokenization is overhyped when people treat it as magic. If the off-chain legal record, custody account, and on-chain token do not reconcile, the blockchain entry is not enough. Real financial market infrastructure needs legal finality, identity controls, sanctions screening, key management, and disaster recovery.

Why AI and Digital Assets Are Converging

AI and digital assets are often discussed separately. Bank of America's structure suggests they are converging inside markets businesses.

AI can help monitor trading behavior, detect anomalies, summarize market commentary, optimize liquidity, and support risk teams. Digital asset platforms can change how instruments settle and how collateral moves. Together, they support more automated markets, but they also require tighter governance.

McKinsey has estimated that generative, predictive, and other AI technologies could create up to $340 billion in annual value for the global banking sector. That value will not come from chatbots alone. It will come from process redesign: fewer manual reconciliations, faster service, better fraud detection, improved credit workflows, and smarter capital markets operations.

What This Means for Financial Innovation

For banks, Bank of America's AI and digital assets expansion raises the competitive bar. AI-first service models will become normal in retail and corporate banking. Tokenized settlement and digital collateral will move from conference panels into controlled institutional rollouts.

For developers, the skill mix is changing. Solidity or smart contract knowledge helps, but regulated finance also requires identity, privacy, audit controls, ISO 20022 messaging awareness, custody models, and secure API design. On the AI side, you need model evaluation, retrieval design, data governance, and prompt testing under compliance constraints.

For professionals, this is a clear learning signal. If you work in banking, payments, compliance, wealth, or capital markets, you will need enough AI and blockchain literacy to question vendors, assess risk, and work with technical teams.

Skills to Build Next

If you want to prepare for this shift, focus on practical foundations rather than hype:

1. Learn how tokenization works for real assets, deposits, and securities.
2. Study stablecoin design, reserve models, custody, and regulatory risk.
3. Understand AI governance, model risk, and data privacy in financial services.
4. Build a small retrieval-augmented generation prototype using approved documents, then test it for wrong answers.
5. Map how a payment moves today before assuming blockchain will improve it.

Relevant Blockchain Council learning paths include Certified Blockchain Expert™, Certified Blockchain Developer™, Certified AI Expert™, and Certified Generative AI Expert™. If your role sits between technology and business strategy, start with blockchain and AI fundamentals. If you build systems, go deeper into smart contracts, secure architecture, and AI application design.

Final Takeaway

Bank of America's expansion shows that financial innovation is moving into a more serious phase. AI is being used to reshape service and operations at scale. Digital assets are being organized around settlement, deposits, collateral, and custody. Your next step is simple: pick one banking workflow, learn how it works today, then study how AI or tokenization could improve it without breaking compliance, trust, or control.