According to a survey conducted by by KPMG, an international consulting and auditing corporation, and researchers from the University of Melbourne, found that 57% of employees surveyed AI-generated content for your own while working.
The survey involved 48 thousand people from all over the world. The results showed that the vast majority of respondents secretly uses AI during operation, and then passes off the results it generates as its own.
In the analysts’ report KPMG’s report entitled «AI Trust» states that a total of 48 thousand 340 people in 47 countries were surveyed between November 2024 and January 2025. 57% of respondents admitted, that hide the use of AI during operation and present the content it generates as their own.
According to the study, 58% of people now intentionally use AI at work, and about a third do so at least once a week. According to professor of management and chair of trust at the University of Melbourne Business School Nicole Gillespie, there is an incredibly high level of opacity in how workers around the world are using AI.
According to her, the hidden use of artificial intelligence systems at work is due to the pressure on employees to use modern advanced tools and meet progressive standards. Gillespie explains that most employees are afraid of falling behind others if they do not use modern tools in their work.
If the management prohibits the use of generative AI models, employees usually start hiding their use. Except people start to see the benefits, they are tempted to continue using AI, even if they know they are violating company policy.
The authors of the study also claim that only 47% of respondents confirmed that they had received training in working with artificial intelligence. This means that the rest simply use AI functions without any instructions.
At the same time 66% of employees said they use AI tools without evaluating the accuracy of their answers, 48% — admitted to uploading company information to publicly available AI tools and 56% — reported making mistakes in their work because of AI.
The researchers also argue that this practice, in addition to errors, data leaks, and risks of non-compliance with regulatory requirements, also undermines trust in AI systems. Experts note that in order to build trust in AI and obtain accurate results from it, it is necessary to ensure better training and management in the field of artificial intelligence.
The study points to an insufficient level of literacy and management systems at workplaces. About half of the respondents admitted that they did not understand AI and how to use it. In emerging economies such as India, Nigeria, and Saudi Arabia, the level of trust in AI was much higher than in developed economies — 82% compared to 65%, which is also due to with a higher level of literacy and education.
Spelling error report
The following text will be sent to our editors: