The Next Frontier of Payments: Embracing Agentic AI
In many ways, artificial intelligence (AI) has become a conventional technology across the payments industry. A year ago, organizations rushed to implement AI solutions; today, this technology is prevalent in several applications.
Earlier iterations of AI often focused on fraud detection and compliance; however, the introduction of generative AI has significantly enhanced these processes. Companies like Mastercard and Swift have adopted this technology in novel initiatives to combat fraud, demonstrating promising results over the past year.
Customer service has also seen improvements, with companies such as Remitly reducing support times by up to 75% through generative AI. Klarna, meanwhile, has fully integrated AI technology to boost operational efficiency, despite recent workforce reductions in advance of its IPO.
The Advent of Agentic AI
Now, the industry stands on the brink of another leap forward with the emergence of agentic AI. Unlike generative AI, which often requires user interaction to produce outcomes, agentic AI operates autonomously. This technology allows AI agents to make complex decisions independently, learning and adapting in real time.
Applications for agentic AI include investment portfolio management, proactive AI-based personal assistants, and virtual healthcare agents. Furthermore, sectors such as DevOps and customer support could benefit from advancements in this technology.
Several startups are developing agentic AI solutions, with major players like OpenAI and Salesforce also entering the space. For instance, OpenAI recently launched Operator, an agentic AI designed to perform repetitive online tasks independently.
Agents in Payments
The potential impacts of agentic AI are being felt in the payments industry. PayPal, for instance, is exploring how this technology can research and make purchases on users’ behalf. Simultaneously, eBay is piloting OpenAI’s Operator to improve access to site listings.
Stripe has begun utilizing agentic AI, launching a checkout solution specifically designed for AI agents. CEO Patrick Collison remarked, ‘Nobody wants a T-1000 going wild on the family credit card,’ emphasizing the importance of control mechanisms in these transactions.
As these early ventures unfold, they may pave the way for a new era in payments involving complete automation—transactions initiated and executed without human input.
The Future Ahead
Agentic AI could significantly expedite transaction processes which have already accelerated with the digitization and movement towards real-time payments. However, a supportive ecosystem of solutions, regulatory frameworks, and technological support is vital for the successful implementation of agentic AI in payments.
While the full potential of this technology remains to be unveiled, its prospects are promising. As leaders in the payments industry explore these innovations, one thing is clear: the landscape of digital transactions is evolving, and those who embrace these advancements will be at the forefront of transforming the industry for years to come.
