When people think about banks adopting generative AI, the conversation usually focuses on customer-facing tools—chatbots answering (or not…) queries, fraud detection systems keeping accounts secure, or personalized recommendations for financial products. But behind the scenes, banks are quietly tapping into the transformative power of generative AI to change how they operate internally. And the changes happening within the walls of these institutions might be even more significant than what customers see on the surface.
GPT Store: Democratizing AI Innovation
BBVA’s ‘GPT store’ is a simple yet transformative idea: a centralized platform where employees can create and access AI-driven tools designed to solve everyday challenges. Whether it’s automating mundane tasks, improving efficiency, or enhancing decision-making, the GPT store effectively ‘democratizes’ AI by putting powerful tools in the hands of employees across departments.
For instance, an employee in compliance might use a GPT tool to summarize lengthy regulatory documents, saving hours of manual effort. Meanwhile, a product manager could leverage the same platform to draft tailored marketing content or analyze customer feedback more effectively. By fostering a culture of innovation and collaboration, BBVA’s approach ensures that AI isn’t confined to IT or data science teams—it’s a resource for everyone. Some might argue that most of what we use generative AI for today is pretty mundane and useless, but it all is likely useful – ‘one person’s trash is another’s treasure’ is just as applicable in the metaverse as it is IRL.
This strategy also reflects a broader shift in how banks are thinking about AI. Instead of relying solely on pre-built solutions from vendors, institutions like BBVA are creating ecosystems where employees can develop AI tools that address their unique needs. It’s a model that prioritizes agility, customization, and ownership, giving employees the power to shape the tools they use daily.
Indeed, much of the coding functionality that we are starting to see in AI obviates the need for many pre-built platforms. Indeed with companies like Klarna effectively pulling the plu on often expensive SaaS systems, many banks are sure to follow.
Empowering Employees with the LLM Suite
JPMorgan Chase has rolled out the ‘LLM Suite,’ an in-house generative AI assistant designed to streamline workflows and improve productivity for its 200k+ employees. From drafting emails to summarizing documents, the LLM Suite tackles repetitive tasks, allowing employees to theoretically focus on higher-value work.
The rollout of the LLM Suite has been accompanied by competition between teams to integrate AI solutions into their processes. JPMC also supports the transition through comprehensive training programs. Employees also have access to “superusers,” experts embedded within teams who assist with AI adoption and ensure seamless integration into daily workflows.
Even JPMorgan’s CEO, Jamie Dimon, is reportedly a user of the LLM Suite, signaling the bank’s commitment to embedding AI at every level of the organization – although one can imagine the queries are slightly different at this point and likely feature: ‘What should I do if Trump does…’
GenAI for Seamless Communication
Morgan Stanley has taken a different approach, focusing on how generative AI can improve internal communication and collaboration. The bank has developed tools like “AI @ Morgan Stanley Debrief,” a generative AI application created in partnership with OpenAI. This tool simplifies meeting management by summarizing video calls and drafting follow-up emails, helping employees stay on top of their workloads.
Morgan Stanley’s strategy revolves around creating custom AI solutions tailored to its workflows. Rather than adopting generic tools, the bank integrates generative AI into its existing processes, ensuring that employees see tangible benefits without having to overhaul their routines. This focus on seamless integration demonstrates how banks can use AI to enhance efficiency without disrupting the way employees work.
Challenges and Risks
While these case studies highlight the potential of generative AI, they also underscore the challenges banks face in implementing these technologies. Data security is a major concern, especially given the sensitive nature of financial data. Ensuring compliance with regulations while using generative AI requires robust governance frameworks and careful oversight.
Another challenge is managing the risks of bias and inaccuracies in AI-generated outputs. Even small errors can have significant consequences for internal tools that influence decision-making. Banks must invest in rigorous testing and validation processes to ensure that their AI solutions meet the highest standards of reliability and fairness.
Finally, there’s the human factor. Adopting generative AI requires a cultural shift within organizations, as employees may be resistant to new tools or concerned about job displacement. Banks need to prioritize change management, providing training and support to help employees adapt to the new technologies and see them as enablers rather than threats. It also begs the question of what roles will be around for these employees in the future, or are they GPTing themselves out of a job?
Will Generative AI Succeed Where Blockchain Didn’t?
As promising as these generative AI initiatives sound, it’s worth asking whether they’ll achieve lasting success—or fade into the background like other hyped technologies. Banks have a history of chasing innovation, only to hit roadblocks when scaling those ideas. Remember blockchain? For years, it was heralded as the next big thing in banking, promising to revolutionize everything from trade finance to cross-border payments. While blockchain has found niche applications, it has yet to deliver the sweeping transformation many envisioned. Much of its potential was stymied by regulatory hurdles, interoperability challenges, and an inability to align expectations with real-world capabilities.
Generative AI could face similar obstacles. While the technology is undeniably powerful, its success depends on more than adoption. Banks must integrate AI seamlessly into workflows, ensure data security, and navigate regulatory complexities. Moreover, they’ll have to prove that these tools deliver measurable value—not just in terms of efficiency but in empowering employees and improving decision-making. Without a clear return on investment, generative AI risks becoming another innovation that looks great on a press release but struggles to impact day-to-day operations.
That said, this time’s key difference may be the technology’s flexibility. Unlike blockchain, which requires entirely new infrastructure and often a rethink of existing processes, generative AI can be layered onto current systems. This adaptability could make it easier for banks to extract value quickly, provided they avoid overpromising and underdelivering. The next few years will be critical in determining whether generative AI becomes a transformative tool or another technology that didn’t quite live up to the hype.
Conclusion
The adoption of generative AI by banks shows that the technology’s most transformative applications may not be customer-facing at all. By focusing on internal operations, these institutions are unlocking new efficiencies, fostering innovation, and redefining how their employees work.
As the financial industry continues to embrace AI, the challenge will be balancing innovation with security, accuracy, and user adoption. But one thing is clear: the real revolution in banking isn’t just about offering smarter services to customers—it’s about building smarter banks from the inside out.
Read the full article here
