Central bankers are some of the most powerful people in the world, and we should all pay attention to what they\u2019re saying and hearing. With that in mind, I have troubling news.<\/p>\n
According to Reuters\u2019 Wednesday report <\/a>from the European Central Bank\u2019s annual meeting in Sintra, Portugal, AI is being talked about like a thousand-mile-wide UFO hovering above Europe. According to Reuters, nearly every time anyone had a conversation, AI came up.<\/p>\n Quotes collected by Reuters from scared and\/or excited central bankers\u2014along with experts who attended the event\u2014are, well, chilling when taken all together. I\u2019ve extracted the choice artifacts so they can be more easily dug up and studied after the dust settles.<\/p>\n Crypto-loving<\/a> U.S. Federal Reserve Chairman Kevin Warsh played the role of cheerleader, or perhaps bubble-inflator. \u201cThis is the biggest time of consequence to each of our economies, I think, in our lifetime,\u201d Warsh said, and added, \u201cWho knew when the internet was born that the internet was going to create a million and a half jobs as Uber drivers? We are in the first to second inning of this revolution.\u201d<\/p>\n Richard Tiffany \u201cTiff\u201d Macklem, Governor of the Bank of Canada \u200balso compared what\u2019s happening now to the rise of the internet\u2014but in a laudatory way. \u201cThe internet proved to be better than anybody imagined, created whole new businesses, but we still got the dotcom bubble,\u201d Macklem said, adding a slight warning to his oddball assessment of one of the worst inventions in human history<\/a>.<\/p>\n And Tobias Adrian, Financial Counsellor and Director of the Monetary and Capital Markets Department of the International Monetary Fund (IMF) fretted about what will happen when banks run agents that make customer-facing decisions. \u201cHow do supervisors assess those kind of agentic loan decisions? They are a little \u2060bit black box. There\u2019s potentially a lack of explainability, and I think that is a key supervisory challenge.\u201d\u00a0Computer has been saying no<\/a> for many years. Does the dawn of the agentic version of computer says no really seem like a major shift? (Honest answer: Maybe).<\/p>\n But according to Reuters, Itay Goldstein, Professor of Finance at the Wharton School of the University of Pennsylvania showed up with bubbles on his mind. If you think we\u2019re in one now, just you wait, Goldstein explained.<\/p>\n \u201cSomething that is even more advanced and potentially more disturbing, is the ability of these algorithms to coordinate on a manipulative path of prices[\u2026]. These algorithms indeed manage to achieve this kind of manipulation, creating bubbles leading to crashes, and this, I think, has more significant implications for financial stability.\u201d<\/p>\n In this context, it\u2019s worth reading what Canadian commentator and novelist Cory Doctorow wrote late last year<\/a> about a bubble. Just after pointing out that LLMs create applications\u2014and what he calls \u201cplugins\u201d\u2014that would be seen as pretty normal in normal times, he made a similar point to Goldstein\u2019s about an AI bubble as an economic force of nature, although he struck a slightly different tone.<\/p>\n The bubble itself, according to Doctorow, \u201cwants expensive, \u2018disruptive\u2019 things: Big foundation models that lose billions of dollars every year. When the AI investment mania halts, most of those models are going to disappear, because it just won\u2019t be economical to keep the data-centers running. As Stein\u2019s Law has it: \u2018Anything that can\u2019t go on forever eventually stops.\u2019\u201d<\/p>\n