A recent study conducted by researchers from Microsoft and Carnegie Mellon University has raised concerns about the impact of generative AI on workplace critical thinking skills. The findings suggest that when employees rely too heavily on AI-generated outputs, they may lose opportunities to practice independent problem-solving, potentially leaving them unprepared for situations where AI falls short.The study, which surveyed 319 individuals who use generative AI at least once a week at work, examined how AI is integrated into daily tasks. Participants commonly used AI for three main purposes: creating content, gathering information and seeking advice. While AI-assisted tasks often led to increased efficiency, the study found that users shifted their focus toward verifying AI-generated responses rather than engaging in deeper analytical thinking.According to the research, only 36 per cent of respondents reported actively applying critical thinking skills to counter potential risks associated with AI reliance. Some participants expressed concerns about AI-generated outputs leading to professional missteps, such as submitting inaccurate performance reviews or sending culturally inappropriate emails. Others sought additional verification from external sources like YouTube and Wikipedia, raising questions about AI’s actual effectiveness in reducing workload.A key insight from the study was the relationship between confidence in AI and critical thinking effort. Participants who trusted AI outputs more tended to exert less cognitive effort in evaluating responses, while those confident in their own abilities were more likely to scrutinise AI-generated content. The researchers stressed that users must understand AI’s limitations in order to mitigate potential risks. Without this awareness, employees may become overly dependent on AI tools, leading to a gradual decline in independent reasoning skills.While the study does not explicitly claim that generative AI makes users less intelligent, it cautions against excessive dependence. The researchers warn that if AI is allowed to take over decision-making processes entirely, workers may lose the ability to think critically when AI inevitably produces flawed or misleading results.