How AI can make financial services smarter and more secure
Implications for banking, insurance, payments, and wealth management
According to the chief customer and digital experience officer for Citigroup's global cards unit, Alice Milligan, “The next generation [of digital] is getting to a place where we’re offering a more predictive and dynamic experience to the customer.”
The modern banking customer is no different from virtually every other modern consumer. They can be perpetually connected to mobile devices and wearables. They increasingly expect personalization that’s seamlessly integrated into daily tasks. Many Financial Technology (FinTech) startups and digital retailers offer consumers this type of connected, customized experience.
For Financial Services (FinServ) providers looking to enhance the customer experience, artificial intelligence (AI) and machine learning can help. These technologies can provide tailored customer recommendations, streamline omnichannel interactions, and support tight integration with services like fraud protection and customer support. What’s more, AI and machine learning can enable these strategies efficiently and cost-effectively.
Inspire renewed loyalty, streamline operations, and position yourself to outpace competitors with integrated, data-driven solutions.
Smart, flexible automation
Introducing AI to a FinServ transaction uses algorithms that can connect the dots between customer behavior and their desired products and services. That means you don’t have to rely on manual assessments. Customer data and analytics can subsequently drive personalization through automation. AI chatbots in a retail bank, for example, can reference both demographic and account-specific information to recommend a relevant offering to a customer while simultaneously running fraud-related evaluations. This automation benefits both the bank and the customer.
AI can unlock the multi-platform potential of cloud services as well as public and private networks. Solutions that rely on network connectivity are becoming more effective when combined with software-defined architectures. Operational benefits are being realized with chatbot technology. It shifts human staffing away from frequent customer service tasks. In addition, as technology accelerates the move to “software-defined everything,” a new era of rapid utilization and development is underway. This is fueling FinTechs and disruptive innovations like blockchain and 5G. AI driven by automation and software-defined orchestration enables a customer journey that’s as much about personal touch as it is about analytics.
For example, purchase approval based on credit history and ATM access—and personalization based on physical location—can be automated using AI. Other use cases, however, require deeper integration with customer journey maps. For users that opt-in, an AI algorithm that monitors spending habits can proactively suggest tailored mortgage services when it detects home purchase activity. This kind of relationship can nurture the perception of the financial institution as a helpful, trusted partner.
“Check my balance” vs. “Should I pay off my loan?”
CB Insights is currently tracking over 7,000 AI deals across all industries, providing a glimpse of how broadly AI is being developed and defined. Most AI use cases wouldn’t receive a starring role in a sci-fi blockbuster, but they’re becoming workhorses behind the scenes. Categories of AI employed for FinServ can be distinguished between Artificial Narrow Intelligence (ANI) and Artificial General Intelligence (AGI). ANI is common today, but AGI is future tech.
ANI is the most broadly-understood use of data analytics and machine learning to drive automation toward a defined outcome using algorithms. ANI has been described as a “one trick pony” because it is often deployed to efficiently accomplish a single task. ANI can be used for tasks related to product or service recommendations, for example, because it excels at reacting to specific datasets and outputting customer analytics and data trends. Think about a self-driving car that knows how to navigate obstacles and respond to traffic conditions. In banking, an ANI-driven chatbot session might help a customer asking routine questions about their account. ANI can enable the chatbot to respond to predetermined key words and phrases, guiding the user through tasks such as “check my balance.”
AGI, in contrast, is referred to as “superintelligence” or “human-level intelligence.” AI researcher Eliezer Yudkowsky describes AGI as an “efficient cross-domain optimization” which has the distinction of evaluating and reacting to multiple environments to deliver undefined outcomes. AGI has also been described as having the ability to “transfer learning” from one domain to another. An AGI self-driving car, for example, could decide to drive off-road in order to avoid a natural disaster at speeds outside the speed limit. It could consider news, weather, and other inputs like social media data to make “human-level” decisions about how and where to take the vehicle to best avoid risk.
For financial service providers, an AGI-driven chatbot might detect a customer’s mood or attempt to infer motive about why a question is being asked in the first place. The customer may query, “Should I pay off my car loan?” Based on this question and available contextual data, AGI could assume the purpose of the question, or it could request more information to confirm the goal. The output presented to the user would continue to be narrowed-down, steering the interaction in a more “human-intelligent” way, minimizing options and information deemed redundant or irrelevant to the conversation. As a result, more complex “contextual advice” could be offered to customers without the wait time frequently associated with these kinds of interactions.
AI can help large organizations minimize perceptions of bureaucracy, instead communicating agility and relevance for competitive advantage.Share this quote
AI can simultaneously introduce and help mitigate new security concerns for FinServ providers. Fraudulent authentication, probing of network vulnerabilities, and social engineering can be potential AI threats that risk nefarious use of data, analytics, and automation. For instance, trends like “deepfakes” are rapidly evolving in capability and effectiveness. A deepfake employs AI to quickly superimpose a customer’s physical attributes, mannerisms, and voice onto unauthorized images, videos, or datasets, posing a potential risk to the integrity of biometric authentication. Yet, at the same time, AI can be deployed to help mitigate the very same risks, as long as the benevolent algorithms outpace the malevolent ones. If that’s the case, the same cybersecurity foundation can often be used to help protect FinServ providers, their alliances, and their customers while also enabling better service.
AI for competitive advantage
Financial institutions that focus more on a strategy that’s tech-agnostic, customer-centric, and future-flexible instead of a specific device, technology, or platform are likely to be better positioned to compete. New analytical and adaptive capabilities can reveal exciting features and benefits that customers are likely to notice. Perceptions of trust and security are critical for FinServ. AI-powered experiences can reinforce those perceptions by surprising and delighting customers with smarter, more relevant interactions. In this way, AI can help large organizations minimize perceptions of bureaucracy, instead communicating agility and relevance as a competitive advantage.
Want to learn more about how AI can help you connect and protect the omnichannel experience? Check out the turnkey solutions in the API Marketplace.
AT&T Business is a leading provider of Edge-to-Edge solutions for Financial Services and is the largest SD-WAN provider globally. Achieve smarter, more trusted interactions with business solutions that integrate our unique ecosystem of technology and expertise with our highly-secure global network to obtain near-real-time intelligence from every corner of your enterprise.