According to a Reuters report on Wednesday, artificial intelligence (AI) dominated discussions at the European Central Bank’s annual conference in Sintra, Portugal. Central bank officials and attending experts presented a range of perspectives on AI’s economic and financial-stability implications.
Optimistic parallels with the internet
Kevin Warsh, the former U.S. central banker noted in the coverage for his enthusiasm toward cryptocurrencies, emphasized potential benefits: “I think this is the most significant time in our lifetimes for all of our economies.” He added, “Who would have thought at the birth of the internet that it would create jobs for 1.5 million Uber drivers? We are only between the first and second acts of this revolution.”
Richard Tiffany (“Tiff”) Macklem, Governor of the Bank of Canada, also likened the moment to the rise of the internet, praising the technology’s long-term capacity to spawn new businesses while cautioning that it also produced the dot-com bubble.
Supervisory and stability concerns
Tobias Adrian, financial counsellor at the International Monetary Fund and director of its Monetary and Capital Markets Department, highlighted oversight issues. He warned about banks deploying autonomous AI agents to make customer-facing credit decisions: “How will supervisors assess such autonomously made credit decisions? These operate somewhat as black boxes. They may lack explainability of decisions, and I think that will be one of the biggest challenges for supervision.”
The report notes that while computers have long been able to deny credit, the rise of autonomous AI agents raises new questions about accountability and the tools supervisors will need.
Bubble risks and market manipulation
Itay Goldstein, a finance professor at the Wharton School of the University of Pennsylvania, warned about economic bubbles. He argued that algorithms could coordinate to manipulate prices, creating bubbles that might end in collapses with serious consequences for financial stability: “There is something even more advanced and potentially more worrying: the ability of these algorithms to coordinate to manipulate price dynamics. If these algorithms can indeed do such manipulation, they could create bubbles that ultimately collapse. I believe this could have far more serious consequences for financial stability.”
Participants also referenced Cory Doctorow’s recent commentary that the AI “bubble” partly hinges on expensive foundation models that burn billions of dollars annually; when the investment mania ends, many of those models could disappear because it will not be economically sensible to run the data centers.
A two-way risk
Torsten Slok, chief economist at Apollo Global Management, summed up the dual nature of the threat: “If artificial intelligence exceeds expectations, it will affect financial stability. If it underperforms expectations, it will also affect financial stability.”
Why this matters
Views voiced in Sintra matter because central bankers’ assessments can shape regulatory, supervisory, and policy responses. AI could provide substantial economic benefits, but novel supervisory challenges, accountability gaps, and model-driven market dynamics pose material risks to financial stability that regulators will need to address.
Participants’ positions ranged from optimistic to cautious to openly concerned, reflecting the mixed and uncertain implications of AI for the financial system.
