Germany’s financial regulator wants to pay closer attention to the banking sector’s artificial intelligence (AI) use.

This move by the Federal Financial Supervisory Authority (BaFin), announced Wednesday (July 29), comes after the watchdog was granted new powers under the European Union’s AI Act.

“We will look into how banks, insurers and other financial entities use AI in direct connection with regulated financial activities,” Jens Obermöller, director-general for cyber risks and technology in the financial sector at Bafin, said in an interview on the regulator’s website. “So we will step in when companies use AI for activities for which they need our authorization—such as for banking and insurance transactions.”

He said that Bafin will examine a sample of the AI applications used by many financial institutions in relevant areas, but does not plan to look into every AI system at every bank, as the AI Act mandates monitoring rather than supervision.

“We monitor whether companies are complying with the transparency requirements and provisions regarding prohibited AI practices,” Obermöller said. “And we’ll be looking at whether companies are taking measures to promote AI literacy among their employees.”

This will happen right away, with Bafin moving to address “high-risk AI” starting in December 2027. Systems that fall under this designation include those used by insurance companies for risk assessment and to price life and health insurance policies.

“AI can, for example, calculate the individual surcharges a customer must pay,” Obermöller said. “In the case of banks and other financial institutions, ‘high-risk AI’ generally refers to AI systems used to evaluate the creditworthiness and credit scores of natural persons.”

In related news, PYMNTS CEO Karen Webster spoke earlier this month with Maik Taro Wehmeyer, co-founder and chief executive at Taktile, who predicted that 2026 will be “the year where AI will come to financial services.”

“The report noted some of the benefits of using AI in commercial lending and insurance. For example, small business loans that once required weeks of manual underwriting could potentially be approved in minutes.” Meanwhile, insurance claims that traditionally needed months of evaluation can be handled within hours with drone imagery and AI-powered damage assessments.

“I think many people by now confuse AI transformation with cost savings,” Wehmeyer said, noting that the bigger competitive advantage now comes from AI’s ability to dramatically shrink decision times.

“If I’m a small business owner and I’m asking for a loan, and I get the answer not within 14 days … but within five minutes, how great is that?” he said.