Artificial intelligence can now be used to keep executives honest at company earnings announcements

Isaac Lawrence / AFP / Getty Images
  • Analytics company S&P Global has developed AI tools to analyse the wording used in company earnings announcements.
  • The technology focuses on language sentiment and language complexity.
  • S&P says the data can provide a useful guide on the outlook for a company‚Äôs stock price.

Artificial intelligence is changing everything — particularly when it comes to gaining unique insights from huge sets of data.

And researchers at analytics company S&P Global have applied the technology to more effective analysis of corporate earnings reports.

S&P’s research used AI techniques and machine learning to focus on two areas — the number of positive or negative words used, and the complexity of language.

Quite simply, they found that if more negative words were used it had a direct relationship with a given company’s earnings outlook.

And they found that language complexity also has negative implications.

The research was based on the view that where companies have negative announcements to report, they often try to avoid the difficult questions by presenting complex, elongated answers.

S&P used AI tools to assess the length of each sentence as well as the number of complex words with more than three syllables.

The technology is able to pick up on longer answers and immediately compare them to the industry average.

The work by S&P Global is reminiscent of ANZ’s Bias Index — a tool the bank developed in August last year to analyse the wording used in the RBA’s interest rate announcements.

The adoption of AI techniques may make it easier for analysts to “see through” the complex language used company executives, thus making it hard to present more negative earnings results without seeing a corresponding decline in the share price.

It could also address another common refrain in financial markets, which is that some companies often lean towards questioning from “favoured” analysts who take a more positive view of their earnings outlook.

The net result is that S&P found AI analysis of language sentiment and language complexity provides a useful guide for the future direction of a company’s share price.

Read more at Business Insider Australia Research >>

Business Insider Emails & Alerts

Site highlights each day to your inbox.

Follow Business Insider Australia on Facebook, Twitter, LinkedIn, and Instagram.