Although agentic AI and quantum computing are still in their nascent stages, financial services leaders already are identifying ways to fuse the two powerful technologies. 

Quantum has the potential to enhance autonomous AI capabilities by “allowing those systems to evaluate a much broader set of scenarios and trade-offs,” BMO Chief AI and Quantum Officer Kristin Milchanowski told FinAi News. 

“Over time, that could mean agents that are not just faster, but more sophisticated in how they manage risk, optimize outcomes and adapt to changing conditions.”

— Kristin Milchanowski, chief AI and quantum officer, BMO

The $1.5 trillion bank has rolled out agentic AI tools in areas including anti-money laundering, wealth management and information retrievals, according to BMO. 

Read part 1 and part 2 of quantum-AI series. 

Combining agentic AI and quantum computing could result in a stronger infrastructure across financial services, Renee Davis, chief executive of tech company OpenMatter Network, told FinAi News.  

OpenMatter Network builds verifiable trust layers to help financial institutions and businesses deploy AI securely.  

When an autonomous agent has access to a quantum computer, “you’re probably going to see faster financial rails, transactions, resource distributions and allocations,” Davis said.  

R&D efficiency  

Agentic AI has recently become “very valuable” for quantum research, Manuel Proissl, global industry applications lead for financial services at IBM Research, told FinAi News.  

Researchers and developers are interacting with AI agents to “study different quantum circuits, different quantum algorithms,” to identify different transpilation processes and maximize hybrid models, he said.  

Transpilation means rewriting an input circuit to match the constraints of a quantum processing unit and then optimizing the new instructions, according to IBM.  

An input circuit before transpilation

A quantum circuit after transpilation

Another way to combine the technologies is by developing an orchestration layer that allows an AI agent to recognize when it needs quantum power to solve a problem, Proissl said. 

For example, a chatbot could be routed to a quantum computer when it needs to resolve a complex query, he said.  

“If I can train an agentic system that helps me in deciding when this problem is worthwhile to involve a quantum computer … then that brings real value,” he said. “This is where I would also see the industry moving toward.” 

Macro shifts 

Quantum computing also could enhance AI agents’ ability to analyze broader market shifts and potential risks for loan portfolio management, Joshua Summers, CEO of agentic lending platform EnFi, told FinAi News.  

“Those loans have risk in aggregate,” he said. “On top of that, the entire U.S. economy has risk related to all the loans against all the different credit types.” 

Although EnFi currently does not need quantum computing for the commercial lending workflows it addresses, that could change if it expands its platform to simulate thousands of risk scenarios on a macroeconomic scale, Summers said.  

Avoiding hype  

The global market for quantum AI is projected to hit $7.8 billion by 2035, up from $280 million in 2025, according to research company Research and Markets.  

Strong quantum-AI use cases in financial services include risk modeling, quantitative investing, capital optimization, portfolio management and fraud prevention, according to previous FinAi News reporting.  

Despite the powerful combination, FIs must start slow and avoid the hype surrounding the technologies, IBM’s Proissl said. 

The idea that an agent can evaluate numerous decision pathways simultaneously by combining it with quantum superposition is one common misconception, he said.  

BMO’s Milchanowski similarly said FIs “need to stay grounded” when it comes to agentic-quantum convergence.  

Register here for the FinAi Lending Summit, set for Oct. 7-8 in Las Vegas.