Francis de Souza, CEO of gene-sequencing company Illumina, tried to keep his speech positive for investors, but artificial intelligence techniques revealed previously unknown emotions during the latest quarterly earnings conference call.
Despite the notable efforts of Illumina CEO Francis de Souza to keep his speech positive for investors, AI techniques revealed previously unknown emotions during his latest quarterly earnings call.
The main focus of discussion at the event was the exciting $8 billion acquisition of cancer screening company Grail. This process has drawn criticism from activist investor Carl Icahn and generated widespread action by competition authorities on both sides of the Atlantic.
During the conference, D'Souza told analysts that the "drama" only affected a small portion of the company's business, but whenever he was asked about Grill, the speaker felt a change in his tone and voice level.
Speech Craft Analytics, which uses artificial intelligence to analyze audio recordings, has revealed changes in the use of oral language, with increased use of “filler words” and hesitation words, such as “em” and “ah,” and even non-words. Verbal.
According to Beechcraft's chief data scientist, David Pope, quoted in the Financial Times, this change in the situation is cause for concern and tension, especially regarding the sensitive issue mentioned. It is noteworthy that De Souza resigned about two months after the event.
Some of the world's most important investors have drawn attention to the new fact that audio recordings can provide advice about the true feelings of executives.
Many investment funds are currently using algorithms to scan conversations at earnings conferences and company presentations for analysis and gathering signals, a process called “natural language processing.”
Investors, with the help of scientists and the support of modern technologies, are trying to extract more information by analyzing the words of messages spoken by executives. This is done using modern technologies such as artificial intelligence.
A research study published earlier this year showed that combining traditional NLP with voice analysis is an effective way to differentiate companies as their disclosures become standardized and take on an almost constant pattern.
