How AI could roil the next economic crisis
Add Axios as your preferred source to
see more of our stories on Google.
Illustration: Sarah Grillo/Axios
The next recession may happen amid a new technological landscape: wide adoption of generative AI, a phenomenon untested by an economic shock.
Why it matters: The potential economic upsides of AI, including how it might fuel productivity growth, are fairly clear. More uncertain are the potential hits to labor and financial markets when the boom time ends.
Driving the news: A top International Monetary Fund official warns the technology might shift a garden variety recession into an economic disaster.
- Treasury Secretary Janet Yellen is expected to echo the risks of AI-powered financial systems in a speech this week, highlighting the "black box" nature of models that are a mystery even to their designers.
What they're saying: "[T]he widespread use of AI could turn an ordinary downturn into a deep and prolonged economic crisis by causing large-scale disruptions in labor markets, in financial markets, and in supply chains," Gita Gopinath, the IMF's No. 2 official, said in a recent speech.
State of play: Gopinath says the globe's experience with automation and labor markets in recent decades offers something of a warning.
- When the economy is solid, businesses are more willing to invest in new technologies and hold onto workers. When the economy slips, companies shed workers to cut costs and might never return to previous employment levels.
- "In the next downturn, AI is likely to threaten a wider range of jobs than in past cycles, including higher-skilled cognitive jobs," Gopinath said, stressing she was outlining a risk, not making a prediction.
- "The result could be unprecedented job losses," with long-term displaced workers who don't have skills fit for an AI-centric economy.
The intrigue: Financial firms that use trading strategies powered by AI might "struggle to respond" to such unprecedented economic conditions, Gopinath said.
- "[T]hey might quickly and simultaneously become overly conservative and rebalance portfolios toward safe assets," Gopinath said.
- As the prices of assets ditched by the models plunge, it could ripple into a "self-confirming spiral of fire-sales and collapsing asset prices across different financial markets."
- Relatedly, supply chain models built on AI could stoke forecasting mistakes that cause huge swings in inventories and production. "This could cause crippling delays and shortages of critical supplies across the global economy," Gopinath says.
The bottom line: "Most international efforts to mitigate AI risk are currently directed toward concerns around security, privacy, ethics, and disinformation," Gopinath said.
- "But we also need a serious international effort to AI-proof the economy."
