Piero Cipollone, member of the Executive Board of the European Central Bank, discussed the implications of AI for the work of central banks at the ‘National Conference of Statistics on official statistics at the time of artificial intelligence (AI)’ in Rome on Thursday 4 July.
Mr Cipollone noted that the use of artificial intelligence (AI) is growing, as are digitalisation and information and communication technologies (ICT). These technologies will change the way the ECB works, particularly in terms of economic analysis, statistics, communication and market and payment infrastructures. However, he felt that technology had not fundamentally changed the ECB’s approach to monetary policy.
He wanted to look at the macroeconomic impact of AI and identified, among other areas relevant to the conduct of monetary policy, three areas in which AI could affect the economy.
Productivity. Mr Cipollone admitted that the potential of AI to boost productivity was undeniable, particularly at the enterprise level. However, he indicated that the final outcome on productivity gains will depend on the speed and scale of AI diffusion across all sectors of the economy.
He also warned against the risk of most of the value created by AI being captured by a few companies, leading to an overly concentrated market structure.
Employment. “Whether AI represents an opportunity or a risk for employment depends on the net effect”, according to Piero Cipollone, indicating that it would also depend on the ability of the workforce to acquire skills complementary to those of AI.
Stability. Finally, he felt that AI could represent opportunities for financial stability, in particular by making risk assessments or capital and liquidity planning more effective. But the economist pointed to risks such as market concentration and conflicts of interest.
Monetary policy. Finally, examining the implications for monetary policy, Mr Cipollone highlighted the uncertainty of AI’s effect on the evolution of upward or downward pressure on prices or on the natural interest rate.
He also pointed out that AI could have an effect on the transmission of monetary policy by creating new economic “winners” and “losers” and, therefore, could change consumer behaviour and access to credit, and therefore affect the response of demand to changes in monetary policy.
He also felt that AI could lead to a change in financial structures, in particular by increasing non-bank intermediation.
Safeguards. Since artificial intelligence is not capable of reflexivity, Piero Cipollone pointed out the need to set limits on AI in order, for example, to protect confidentiality and privacy or prevent it from facilitating the dissemination of false information, fraud and cyber attacks.
Finally, he warned against over-reliance on AI.
In his view, the judgement, responsibility and decisions of central banks should remain a human responsibility.
Link to the speech: https://aeur.eu/f/cyf (Original version in French by Émilie Vanderhulst)