Deutsche Bank’s New Tool Uses AI to Generate Financial Decisions From ESG Reports
Deutsche Bank’s Data Innovation Group has developed an artificial intelligence tool that showcases one possible use of natural language processing in the financial industry.
Developed by data scientists at the bank, the AI, called Alpha-Dig, analyses companies’ Environmental, Social and Governance reports, allegedly cutting through so-called "greenwashing" - the positive spin and convoluted language characteristic of most companies’ reports. The AI has been trained on data from 1000 organisations to cut through to the non-financial news that influence investment decisions. It uses NLP to infer the context of a typical ESG report and generate 'buy' and 'sell' signals accordingly.
The bank claims that in the years since the 2009 financial crash, the average number of words per quarterly SEC filing has grown by about 20,000 words, while annual filings have swelled to 50,000 words on average. Deutsche believes that the tool addresses a real need in the market, as traditional systems that aggregate keywords and then assign scores to stocks are becoming outdated "yet investors persist with these systems as a human simply cannot read the millions of pages of information available".