European Central Bank Explores Artificial Intelligence for Better Understanding of Inflation

JAKARTA - The European Central Bank (ECB) is considering how to use artificial intelligence (AI) to improve its understanding of inflation. This came after years of downplaying price pressures and delaying the start of the most aggressive policy in its history.

Joining many other companies already using AI, ECB is now exploring ways to process and analyze millions of data points. It includes public price data, company statistics, news articles, and bank supervisory documents to provide better analysis for policy decision making.

"AI offers new ways for us to collect, clean, analyze, and interpret this wealth of available data, so that insights can feed into work in statistics, risk management, banking supervision, and monetary policy analysis," reads a blog post that published by the ECB on Thursday, September 28.

The ECB has for years underestimated inflation, and some policymakers have even openly questioned its models and the sustainability of pursuing policies based on figures that always require upward revision.

One of several AI initiatives, the bank wants to deepen its understanding of pricing behavior and inflation dynamics.

Using web scraping techniques, the ECB can collect a lot of real-time price data, but the numbers are unstructured and not suitable for calculating inflation. Therefore, the ECB wants to leverage AI to organize data and improve its analysis.

Another initiative is to automate the process of classifying data from tens of millions of companies, banks and public sector entities, so that they have a better understanding of their financial condition.

The ECB also hopes to use AI to simplify its communications, countering criticism that its overly complex and technical language is difficult for ordinary people to understand.

"A large language model could also help improve texts written by staff members, making ECB communications easier to understand by the public," the bank said.

"Relatedly, we have been using neural network machine translation for quite some time now to help us communicate with European citizens in their native languages," the bank said.