Major US companies like Starbucks, Walmart, and Delta Airlines are using AI firm Aware to monitor employee messages. The tool scans platforms like Slack and Microsoft Teams for keywords indicating dissatisfaction, safety risks, or unionisation efforts. Privacy concerns arise as the AI flags messages for management review.
A growing number of major
companies
in the US are turning to artificial intelligence (AI) to monitor
employee messages
, sparking concerns about
privacy
, freedom of speech, and potential misuse. Companies like Starbucks, Walmart, and Delta Airlines are reportedly using Aware, an
AI
firm specialising in analysing employee sentiment and potential threats from workplace communication.
According to a detailed report by CNBC, Aware’s software scans internal communication platforms like Slack and Microsoft Teams, looking for keywords and phrases that could indicate dissatisfaction, safety risks, or even unionisation efforts. The tool then flags these messages for management review, raising a red flag for many employees and privacy advocates.
Jeff Schumann, the co-founder and CEO of a Columbus, Ohio-based startup, told CNBC the significance of AI in enabling companies to gauge the pulse of employee sentiment in real-time. Rather than relying on sporadic surveys conducted annually or biannually, Schumann emphasises the role of AI in helping organisations comprehend the risks inherent in their communications dynamics.
According to Schumann, Aware's analytics tool, designed to
monitor
employee sentiment and toxicity, prioritises privacy by abstaining from flagging individual employee names. However, in cases of severe threats or predetermined risk behaviours identified by the client, the company's separate eDiscovery tool is equipped to identify specific employee names.
Aware's data repository encompasses a vast corpus of 6.5 billion messages, reflecting over 20 billion individual interactions across more than 3 million unique employees, as per the CNBC report.
Upon the onboarding of a new client, Aware's AI models undergo a training period of approximately two weeks. During this time, they analyse employee messages to discern patterns of emotion and sentiment within the company. This training enables the AI to differentiate between normal and abnormal communication patterns, Schumann added.
Interesting, Schumann claims that the employees' names aren't revealed to the company. “It won’t have names of people, to protect the privacy,” Schumann said. Schumann told CNBC, clients will see that “maybe the workforce over the age of 40 in this part of the United States is seeing the changes to [a] policy very negatively because of the cost, but everybody else outside of that age group and location sees it positively because it impacts them in a different way.”