A new survey shows that one in three workers are keeping secret the fact that they use artificial intelligence (AI) tools at work. According to a study by Deloitte, which surveyed 25,000 UK workers, two-thirds have already used generative AI—software that can create text, images, or other content—to help with their jobs. Nearly a third of those who use AI say they do so without their employer’s knowledge. This practice is called "shadow AI," where employees use their own AI tools, often without official approval, to complete tasks more efficiently. Online forums like Reddit have sparked discussions about the real-world effects of this trend. One user shared that a coworker was using ChatGPT, a popular AI tool, to complete 80% of their work and was being praised for their efficiency. This raised the question of whether to report the coworker to management. Another user mentioned they had completed two weeks' worth of work in just an hour using AI and worried about the consequences of revealing their AI use. Some users suggested editing AI-generated content to make it sound more human and less likely to be detected. While shadow AI doesn’t always mean workers are acting with bad intentions, it often reflects a preference for tools they are familiar with over those provided by their employer. Many workers say they use AI tools they trust more than company-approved options, especially when their employer hasn’t provided any AI solutions. However, using unapproved AI chatbots can pose risks, as highlighted by a PagerDuty survey, which found that 88% of workers have shared work-related information with public AI systems—such as emails, meeting notes, and even sensitive business data. Workers also express a lack of trust in their employers’ understanding of AI. A PagerDuty survey found that 72% of workers believe they understand AI better than their company’s tech teams, and 77% feel that AI restrictions imposed by their employers are hindering their professional growth. This creates a challenge for companies, who must balance the potential productivity gains of AI with concerns about data security and fairness in the workplace. Additionally, some employees are resistant to using AI tools, with one Reddit user asking how to tell their boss they are against using ChatGPT for basic tasks. According to the Deloitte survey, UK workers are spending around £958 million each year on generative AI tools for work, with nearly a third of them not telling their employers. This large investment raises questions about the long-term effects of such practices, especially when workers are using AI to boost productivity without their employer's awareness. As AI becomes more common in the workplace, companies are struggling to create policies that keep up with its rapid adoption.