Deloitte survey finds nearly a third of UK generative AI users hide it from employers

The TechRadar article cited Reddit discussions to illustrate the issue. One user complained that a coworker was secretly using ChatGPT to do 80% of his work and was being praised for efficiency, and asked, "Do I tell the bosses?" Another user wrote that they had done about two weeks' work in an hour and asked whether to tell their boss or keep it secret. A popular response advised: "Remember, AI work is more detectable than you think. Spend time humanizing the output and adding your own texture, then shut up."

Shadow AI does not necessarily mean employees are acting maliciously, according to the report. In many cases, workers prefer AI tools they already know rather than whatever their company has provided, if it has provided anything at all. But the article drew a distinction between using a favorite productivity app and using an unapproved AI chatbot.

Citing a June report by TechRadar colleague Craig Hale, the article said a PagerDuty survey found that 88% of workers had shared work-related information with public AI systems. About 43% had uploaded emails or correspondence, 40% had shared meeting notes, 34% had entered customer information, and 31% had shared sensitive business information such as financial data or confidential company documents. That risk is where shadow AI becomes more than an awkward workplace conversation, because employees can feed sensitive company information into AI services employers have not approved or secured.

Workers also do not necessarily trust their employers to make the right decisions about AI, the PagerDuty survey found. Seventy-two percent of workers believed they understood AI better than their own tech teams, while 77% thought their company's AI restrictions were limiting their professional development, according to the TechRadar report.

The issue is complicated by deep divisions over AI. The TechRadar article cited a Reddit user asking: "How do I tell my boss who keeps telling me to use ChatGPT for simple tasks that I'm anti-AI?" Workplace rules have not necessarily caught up with the speed at which employees have adopted AI.

Deloitte's research also estimated that British workers are spending £958 million of their own money every year on generative AI tools they use for work, according to the report. That is a significant sum for employees to spend on tools designed to make them better at jobs someone else pays them to do, particularly when almost a third are not telling their employers they use them. TechRadar noted that if a worker has just discovered a tool that lets them do two weeks' work in an hour, telling a boss might lead the boss to wonder what the worker will do with the rest of the week.

Editor's Summary The Deloitte survey shows widespread workplace adoption of generative AI alongside significant undisclosed use, while PagerDuty data points to sensitive information being shared with public AI systems. The findings highlight tensions between employee productivity, employer oversight and data security as workplace rules lag behind AI use. Nearly a third of generative AI users are hiding the tools from employers, and British workers are spending an estimated £958 million annually on their own work-related AI tools.