A mature AI strategy isn’t just a nice-to-have—it’s table stakes, driving insights that transform customer experiences, boost operational efficiency, and maximize profit margins.
That said, many banks still struggle to turn their vast stores of data into actionable outcomes, even if they have done the work of breaking data silos and creating a centralized data repository.
The good news is that traditional banks don’t need to completely forge their own path. Several financial institutions have embraced a tech-forward yet customer-centric mentality that pushes AI capabilities forward without overextending risk.
Here is how Citigroup, JPMorgan Chase, and Bank of America evolved their AI strategy to unlock growth, streamline processes, and build trust in an increasingly data-driven economy.
Citigroup
This American multinational has always prided itself on innovation, taking major strides forward in the 20th century by introducing ideas like compound interest on savings and customer checking accounts. It’s no exaggeration that Citigroup’s novel thinking has repeatedly reshaped the financial services industry. Now, they’re embracing AI and data-driven strategies that are overhauling our world.
Citi guides their digital strategy with three core principles:
- Improving the digital experience to assist customers in getting what they need.
- Ensuring customers have a simple and satisfying overall experience.
- Connecting digital and analog channels to offer 360-degree, data-driven relationships.
By putting this ethos ahead of artificial intelligence solutions, Citi is better equipped to offer customers what they really want rather than just checking a box.
Additionally, Citigroup, the parent company, rolled out artificial intelligence tools in 2024 to support 140,000 employees in eight countries:
- Citi Assist – This internal tool is designed to help employees navigate internal bank policies and procedures. Employees can ask questions and receive answers about HR, risk, compliance, and finance stances.
Creating this type of self-service central repository for critical information can increase efficiency and streamline knowledge sharing. Whether onboarding new employees or providing answers for seasoned team members, this AI tool has the potential to elevate their organization.
- Citi Stylus – Another internal tool, Stylus offers employees the ability to summarize, compare, or search multiple documents simultaneously. This tool has the potential to reduce the time spent on manual processes or switching between documents and systems. Definitely a win for employee productivity.
More than just setting these tools and forgetting them, Citigroup plans to monitor how employees interact with these tools to identify new applications that can further simplify work and increase productivity. Auditing user adoption, potential biases, and functionality gaps can benefit any organization.
JPMorgan Chase
Few financial services organizations have invested in data strategy and artificial intelligence like JPMorgan Chase. As a rule of thumb, they take a pragmatic but disciplined approach to innovation, refusing to pursue flashy applications in favor of profitable pursuits.
In fact, their proactive AI research is pushing the boundaries not only of what their organization can achieve but how the broader business community can harness the technology. Here are just a few of their recent examples:
- Exploring Discrete Denoising Models – Maintaining privacy is always a concern for financial services organizations when using AI models. The potential for misuse or theft can put customer information and regulatory compliance at risk.
However, JPMorgan is exploring a way to generate synthetic data that preserves privacy while also incorporating a small amount of noisy data, mirroring the extra data points that exist as outliers or incorrect entries that are inherent in real-world samples. Creating these natural fluctuations allows synthetic data to reflect real scenarios.
- Training Fair Machine Learning Models – Unintentional biases in AI models, especially when it pertains to demographic subgroups like race and gender, is a common concern in the fight for equity and accuracy. Fair machine learning models are a potential solution, but require exposing models to authentic sensitive data or building hyper-parameters that aren’t practical.
JPMorgan Chase is proposing a framework called Antigone, which generates pseudo-sensitive attributes and uses them as stand-ins for actual sensitive attributes within models. Theoretically, these proxy labels can be used to minimize the gap in fairness with AI analysis, which could help to reduce bias in everything from risk assessment to loan approvals.
All this AI research and use cases tie into their core data strategy. Their commitment to exploring the frontier of artificial intelligence, founded on a robust data strategy, is providing incredible insight that organizations should model. When guided by AI-savvy employees or solutions partners, they’ll have the potential to promote unprecedented change through both algorithmic innovation and systematic data governance.
Bank of America
As an early adopter of AI-driven solutions, Bank of America has had some time to explore the potential of this technology while troubleshooting some of the little problems. Erica, their AI-powered virtual assistant, was the first chatbot to help retail and corporate banking customers with various tasks. The market has since followed in their footsteps.
The AI strategy implemented by BofA is guided by some clear principles that should be used as an industry-wide policy:
- Human Oversight – Encouraging experts to review AI decisions ensures ethical decisions and mitigates risks.
- Transparency – Providing means of understanding how AI systems make decisions so organizations can correct biases and mistakes before they cause harm.
- Accountability – Making leaders accountable for the actions of artificial intelligence not only creates ownership when things go wrong but provides trust that fixes will be made.
Again, AI strategies are not only shaping the customer experience, but also internal operations and efficiencies. A recent press release shows how Bank of America is widely implementing this technology across their organization.
- Erica for Employees – Like Citi Assist, Erica for Employees offers in-house staff answers to health benefits, payroll, and other policy questions. Yet this virtual assistant goes a step further. Erica for Employees is also a rapid tech support tool, guiding staff on password resets, device activation, and other troubleshooting activities.
BofA has experienced some early success with Erica for Employees, reducing IT-service-desk calls by more than 50%. IT staff can then focus on more complicated issues and non-technical staff can surmount roadblocks.
- The Academy – Ongoing learning is vital to any organization’s success and BofA has harnessed AI to help with professional development. The Academy simulates conversations as an interactive coaching tool, giving employees real-time feedback that can boost their proficiency.
In 2024 alone, Bank of America’s employees completed over one million simulations that have helped them to improve client conversations and deliver more consistent services.
- Coding Assistance – Bank of America is also exploring ways to empower their team to accelerate and improve the software development lifecycle. They are currently using a GenAI-based tool to collaborate with code writing, QA, and optimization.
This tool has provided them with efficiency gains over 20%. As long as people are involved in the process of verifying code functionality and interoperability, this use case is a great way to boost development.
These are just a sampling of the AI use cases that Bank of America is implementing. The takeaway from across these AI initiatives is that they satisfy what employees and customers need and want. End users can quickly get answers, receive real-time feedback, and complete routine tasks with greater efficiency.
Building a Better Financial Services AI Strategy
The success stories from JPMorgan Chase, Citigroup, and Bank of America provide valuable insights for financial institutions looking to enhance their AI strategies. As you’re implementing your own artificial intelligence solutions, follow these key takeaways from industry leaders:
- Prioritize Practical Value Over Hype – Focus on initiatives that solve real business problems rather than chasing trendy applications. You’ll get a better ROI and be able to justify future investments.
- Put People First – Center AI strategy on improving experiences for both customers and employees, addressing genuine pain points. And if end users aren’t getting value, quickly identify the gaps in their experience.
- Establish Strong Governance – Your brand’s reputation and trust are on the line with every AI initiative. Be sure to implement human oversight, transparency, and accountability to ensure responsible AI use.
- Balance Innovation with Risk Management – Advance AI capabilities while maintaining regulatory compliance and data security. Blindly implementing solutions without calculating the liabilities is a way to risk fines and lawsuits.
By adopting these principles, financial institutions can transform AI from a mere technology initiative into a strategic capability that drives meaningful outcomes. That way, you can do more than keep a seat at the table – you can walk away with more of the chips.
Ready to revamp your financial services AI strategy? w3r Consulting is here to help build a data-driven, AI-forward organization.
Discover our AI-enablement expertise
Related Articles
Why Every Financial Institution Needs a Robust Merger and Acquisition Data Strategy
Overcome Your 2024 Banking Industry Challenges with Smarter Data Strategies