The European Central Bank (ECB) is looking to use artificial intelligence (AI) to address a long-standing issue: underestimating inflation and its repercussions on monetary policy.
After years of grappling with this challenge, the ECB has recognized the potential of AI to enhance its understanding of inflation dynamics, streamline its operations, and communicate more effectively with the public.
The European Central Bank (ECB) headquarters building is pictured ahead of the start of the press conference on the eurozone's monetary policy in Frankfurt am Main, western Germany on February 2, 2023. (Photo by Daniel ROLAND / AFP)
Underestimating Inflation
For years, the ECB has been haunted by a critical blind spot: underestimating inflation.
Reuters reports that this has delayed the implementation of vital policy-tightening measures, casting doubts on the reliability of the bank's models and decision-making processes.
The repercussions have been significant, and policymakers have openly questioned the ECB's ability to base sound policy on constantly revised data.
AI to the Rescue
Recognizing the urgent need for a solution, the ECB is turning to AI, the bank highlighted in a blog post. The central bank is exploring ways to harness AI's power to process and analyze various data sources, including public price data, corporate statistics, news articles, and bank supervisory documents.
These sources, though abundant, present a challenge due to their unstructured nature.
Through AI-driven techniques, such as web scraping and machine learning, the ECB aims to transform this unstructured data into actionable insights.
By automating data classification and structuring, the bank can better understand the financial state of millions of firms, banks, and public entities.
This move promises to enhance the accuracy and efficiency of the ECB's analyses, particularly in the context of inflation dynamics.
Read Also: AI-Generated Image of 'Tank Man' Surfaces as First Result in Google Search
Athena Platform
One area where AI is already making a substantial impact is banking supervision. The ECB has developed the Athena platform, which employs natural language processing (NLP) models.
This platform assists supervisors in analyzing diverse text documents, including news articles, supervisory assessments, and bank documents. It facilitates topic classification, sentiment analysis, dynamic topic modeling, and entity recognition.
What's striking about Athena is its ability to collate enriched texts within seconds, providing supervisors with the relevant information they need without the tedious process of manual searching.
Generative AI Trials
Financial Times reports that ECB's "infinity team" has launched nine trials of generative AI, a technology capable of creating sophisticated text, images, and even computer code.
These trials encompass various domains, from generating first drafts of briefings and summarizing meetings to drafting software code and translating documents.
The aim is to leverage AI to improve efficiency and quality in various aspects of the ECB's operations.
Myriam Moufakkir, the ECB's new Chief Services Officer, has voiced cautious optimism about AI's potential. She acknowledges the need to address concerns about reliability, legal risks, and transparency.
AI's Role in Communication
The ECB also recognizes the power of AI in simplifying communication. Criticized for its overly complex, technical language, the bank aims to use large language models, such as ChatGPT, to improve the clarity of its messages for the public.
These models can help draft summaries and briefings, and even enhance staff-written texts, making the ECB's communication more accessible to ordinary citizens.
Stay posted here at Tech Times.
Related Article: MIT Merges Physics and AI for Image Generation With Poisson Flow Generative Model++
