AI agents go well beyond traditional chatbots
AI-powered customer service tools like chatbots that rely on large language models (LLM) aren't new: 41% of organizations use AI-powered copilots for customer service and 60% have implemented them for IT help desks.[1]
Perhaps one of the most famous chatbots is Bank of America's Erica®, launched in 2018 as the first widely adopted AI-driven virtual financial assistant. Seven years later, customers have interacted with Erica more than 2.5 billion times, with 20 million clients now actively using the virtual assistant.[2]
Klarna also made headlines when it revealed in 2024 that its generative AI-powered chatbot had 2.3 million conversations with customers in the first month of launch, doing the work of 700 full-time agents. Customer satisfaction was on par with human agents.[3]
Chatbots are getting better and better at answering simple customer queries. But they are ubiquitous and don't differentiate one financial institution from another. However, generative AI has the capability to ingest even more vast amounts of data from multiple data sources and third-party applications. Financial institutions can use generative AI to buildAI agents that deliver sophisticated reasoning and solve complex, multi-step customer problems. The AI agent can autonomously perform certain defined tasks, such as reconciling financial statements or drafting detailed responses to customer questions.
"These AI agents are now making employees more productive, delivering more personalized services in real time, and automating functions to reduce costs," shared Malcolm DeMayo, Global Vice President - Financial Services Industry at NVIDIA. "As a reference point, Bank of New York shared in their 2024 annual results that they experienced half a billion in value creation due to AI."
What is the key difference between the older chatbots and an AI agent? A customer opens a chat on your online banking site and asks for their credit card balance. The chatbot answers quickly and accurately. But an AI agent could not only check the balance but analyze the customer's full financial picture across different accounts and institutions and then deliver suggestions on how to pay off the credit card.
That's a differentiated customer experience.
Feeding Agentic AI
According to NVIDIA's 2025 State of AI in Financial Services report, six out of ten (60%) financial institutions say that their top use case for generative AI is customer experience and engagement, up significantly from 25% in 2023.[4] How they address these use cases is evolving based on generative AI technology advancements.
For instance, financial institutions can combine generative AI with an avatar personality that uses natural facial expressions and body language to respond to customers in a more human-like manner.
Creating effective AI agents that can independently analyze challenges, develop strategies, and execute tasks requires four key building blocks.
- Increasing volumes of data: Financial institutions need to collect and organize customer data to provide context-aware responses to both employees and customers.
- Memory functions: Remembering past customer interactions and using those to respond to customers feels more human-like.
- Continuous improvement: Financial institutions should review AI agent responses and update as necessary.
- Retrieval-augmented generation (RAG): Enhancing the accuracy of generative AI models enables AI agents to deliver personalized and sophisticated responses.
AI Agents Become More Human
Sentiment analysis, which analyzes text to determine the emotions behind the interaction, is also humanizing AI agents. A human can interpret whether a customer is frustrated or angry and adjust their response to the customer's frame of mind. Likewise, an AI agent can analyze emails, chat transcripts, and social media comments and reviews to determine the customer's attitude and respond accordingly.
AI agents can enhance customer experience, but they won't replace real humans. Customers love the convenience and speed of AI chatbots, but customers want options, with 75% of consumers saying they prefer talking to a real human in-person or over the phone for customer support.[5]
However, the more human-like and nuanced AI agents become, the more reliant customers will become on AI agents. Even still, financial institutions need to remain human-centric, especially for emotionally fraught transactions such as buying a first home or investing for retirement.
At bank branches, employees focusing on complex and high-stakes financial decisions can be augmented by AI agents at kiosks that perform automated tasks such as scheduling appointments or even providing a primer on a financial literacy topic.
Here are just a few of the top customer experience use cases for generative AI:
- Analyzing customer conversations in real time and providing recommendations to the agent.
- Enabling people with hearing or speech impairments to consume content more easily.
- Offering engaging interactions with live captioning and human-like voices.
Creating More Personalized Experiences
Creating more personalized customer experiences is an opportunity for financial institutions, and most want to move quickly. The
"The Dell AI Factory with NVIDIA provides the infrastructure and well-defined framework, including LLMs, to enable IT teams to quickly build, test, and deploy AI applications," added Ricardo Tavares, Director, Global Industries FSI Vertical, Dell Technologies.
About the Dell AI Factory with NVIDIA
To Dell Technologies and NVIDIA can help you leverage AI to drive innovation and achieve your business goals. The Dell AI Factory with NVIDIA is the industry's first and only end-to-end enterprise AI solution*, designed to speed AI adoption by delivering integrated Dell and NVIDIA capabilities to accelerate your AI-powered use cases, integrate your data and workflows, and enable you to design your own AI journey for repeatable, scalable outcomes.
* Based on Dell analysis, July 2024. Dell offers solutions with NVIDIA hardware and software engineered to support AI workloads from PCs with AI-powered features and workstations to Servers for High-performance Computing, Data Storage, Cloud Native Software-Defined Infrastructure, Networking Switches, Data Protection, HCI and Services.
[1] https://www.nvidia.com/en-us/use-cases/ai-assistants/
[2] https://newsroom.bankofamerica.com/content/newsroom/press-releases/2025/02/digital-interactions-by-bofa-clients-surge-to-over-26-billion--u.html
[3] https://www.klarna.com/international/press/klarna-ai-assistant-handles-two-thirds-of-customer-service-chats-in-its-first-month/
[4] https://resources.nvidia.com/en-us-2025-fsi-survey/ai-financial-services
[5] https://www.five9.com/news/news-releases/new-five9-study-finds-75-consumers-prefer-talking-human-customer-service