Artificial intelligence is poised to transform the financial sector and broader economy, creating opportunities and risks that central banks must navigate carefully, according to a top official at the Bank for International Settlements.
In a Sunday (June 30) speech at the BIS annual meeting, Hyun Song Shin, economic adviser and head of research, explained how AI could enhance central banks’ ability to monitor economic trends and detect financial crimes while potentially amplifying market volatility and cyber threats.
AI “has taken the world by storm and set off a gold rush across the economy, with an unprecedented pace of adoption and investment in the technology,” Shin said.
Shin highlighted how modern AI systems excel at finding patterns in vast amounts of unstructured data, making them well-suited for applications in finance and economics. Unlike previous narrow AI systems designed for specific tasks, the latest large language models are versatile “zero-shot learners” that can tackle unfamiliar problems with minimal additional training.
According to Shin, this versatility stems from their training on “the totality of the text and non-text data on the internet.” As a result, “AI has moved from narrow systems that solve specific tasks to more general systems that deal with a wide range of tasks, and all in ordinary language rather than in specialized code.”
For central banks, an application could be in economic forecasting and “nowcasting” of current conditions. Shin suggested AI models can combine traditional time series data with nontraditional sources like satellite imagery and social media posts to produce more accurate and timely estimates of economic activity.
However, he cautioned that “central banks should not succumb to ‘magical thinking’ — that somehow the tools alone will bring miraculous outcomes. Timely and plentiful data are key to the success of nowcasting applications.”
Shin said the most promising areas for AI in central banking are payment systems and the detection of financial crimes. He cited a BIS Innovation Hub project called Aurora, which found that machine learning models “outperform the traditional rule-based methods prevalent in most jurisdictions” for identifying money laundering networks.
The project demonstrated that “machine learning methods really excel when data from different jurisdictions are shared in a privacy-preserving way,” Shin noted. “Data cooperation improves detection dramatically over the current rule-based method.”
However, AI also poses risks that central banks must address.
“Reliance on the same handful of algorithms could amplify procyclicality and market volatility by exacerbating herding, liquidity hoarding, runs and fire sales,” Shin said regarding financial stability.
Cybersecurity is another key concern, as AI could enable more sophisticated attacks. At the same time, most central banks surveyed by the BIS said they believe AI can enhance cyber defenses, particularly for “automation of routine tasks or threat detection,” Shin said.
Looking at the broader economic impacts, Shin said the effects of AI will depend on how many workers it displaces, how much it boosts productivity, and how many new jobs it creates. While AI will likely increase overall economic output, its near-term impact on inflation is uncertain and will hinge on whether it stimulates demand more than supply.
The assessment comes as central banks worldwide grapple with persistent inflation and the aftermath of aggressive interest rate hikes. The potential for AI to disrupt labor markets and productivity growth adds another layer of complexity to monetary policy decisions.
To address the myriad challenges posed by AI, Shin called for greater cooperation among central banks.
“The pooling of resources and knowledge can mitigate resource constraints and lower the barriers for central banks in using AI tools,” he said.
Specifically, he suggested central banks would benefit from sharing specialized AI models, comparing notes on policy issues, and collaborating on data production and governance.
“There is an urgent need for central banks to come together to foster a ‘community of practice,’” Shin said.
The speech underscores how AI has emerged as a critical issue for monetary policymakers and financial regulators. While the technology offers new tools, it also threatens to upend existing economic relationships and amplify financial system vulnerabilities.
As they seek to harness AI’s potential while mitigating its risks, central banks must invest heavily in new capabilities and forge closer partnerships with each other and the private sector. Shin’s call for a “community of practice” suggests the BIS sees international cooperation as essential for keeping pace with AI’s breakneck progress.
“We should not underestimate the efforts needed to harness the full potential of AI,” Shin concluded. “But the fruits of cooperation in a community of practice will be considerable, and the BIS stands ready to play its part.”
This push for collaboration comes as central banks face increasing scrutiny over their ability to adapt to changing technological and economic landscapes. By working together on AI initiatives, they may be better positioned to stay ahead of the curve and maintain their effectiveness in an increasingly digital financial world.
For all PYMNTS AI coverage, subscribe to the daily AI Newsletter.
We’re always on the lookout for opportunities to partner with innovators and disruptors.
Learn More