OSFI: AI and cybercrime now top risks to Canadian banks

Artificial intelligence and reliance on third-party service providers among the greatest challenges facing financial institutions

OSFI: AI and cybercrime now top risks to Canadian banks

Canada’s top banking regulator has flagged artificial intelligence (AI), cybercrime, and reliance on third-party service providers as growing threats to the country’s financial system.

The Office of the Superintendent of Financial Institutions (OSFI) said these threats, along with sophisticated financial crimes such as money laundering and fraud, are putting increased pressure on the country’s top lenders.

In an update to its Annual Risk Outlook, OSFI warned that AI can exacerbate existing risks such as cybersecurity breaches, third-party vulnerabilities, data privacy issues, potential bias in decision-making, and even how criminals may exploit these technologies for money laundering and fraud.

“The dominance and global reach of some third-party service providers creates concentration risk that could cause a third-party incident to affect multiple institutions at once as well as exposure to incidents outside of Canada,” OSFI said.

Canadian financial institutions rely on a “complex network” of third parties, which OSFI sees as a potential weak spot in the face of growing cyber threats.

The lack of “robust regulatory frameworks” in some third-party industries further increases the vulnerability of these institutions, making them more prone to data breaches and cyberattacks.

“Cyber incidents continue to accelerate across all industries, affecting Canadian institutions and their third-party networks,” OSFI noted, highlighting that these risks are evolving rapidly.

How OSFI is responding

To address these growing risks, OSFI plans to assess how well institutions are prepared to manage their third-party relationships, technological risks, and cyber threats. The regulator will also strengthen its guidelines on AI adoption to reduce related risks and gather third-party data to monitor how effectively financial institutions are managing cybercrime threats.

While AI has its dangers, Canadian banks are fully embracing its potential. Royal Bank of Canada, Toronto-Dominion Bank (TD), and Bank of Montreal (BMO) ranked within the top 10 globally for AI research, according to an April report by Evident Insights Ltd.

TD, for example, is already using AI to pre-approve customers for mortgages and lines of credit. It also provides a virtual assistant to help customer service agents with day-to-day tasks.

The Bank of Canada has also adopted AI for tasks such as forecasting inflation, tracking the economy, and verifying regulatory data.

“We’ve only just begun to explore this technology,” said Bank of Canada governor Tiff Macklem in a recent speech. “With very large and highly disaggregated data sets now available, there is huge potential to use AI to understand how consumers and businesses are behaving and how companies are setting their prices.”

Meanwhile, BMO has taken its AI efforts further by appointing Kristin Milchanowski as its new chief artificial intelligence and data officer. Milchanowski, who completed her postdoctoral studies in AI at the University of Oxford, brings over 20 years of experience in the financial sector and will focus on driving the bank’s “culture of innovation and accountability.”

Why the update?

OSFI typically releases its risk outlook once a year in the spring. However, with risks continuing to evolve, the regulator has added new areas of concern. Back in May, OSFI flagged real estate lending, wholesale credit, funding and liquidity, and geopolitical risks as top concerns.

But now, AI and cyber-related threats have joined that list.

Read next: Embrace AI in mortgages: adapt or be left behind

“AI has very significant risks, but also significant opportunities,” said OSFI’s superintendent Peter Routledge at the Global Risk Summit on Wednesday.

He noted that regulatory responses to AI may take time to implement, which raises the risk of AI evolving so quickly that it creates unforeseen challenges.

“AI could really take off and create risks we don’t see and create a crisis that’ll cause us to react to it,” Routledge added.

Make sure to get all the latest news to your inbox on Canada’s mortgage and housing markets by signing up for our free daily newsletter here.