Inside the Growing Problem of AI Misuse in US Companies
500It turns out that AI misuse isn’t just a problem for employees in lower roles. A new report from AI security firm CalypsoAI shows that even top executives are playing fast and loose with AI rules. The study highlights a surprising trend: many in the C-suite are adopting AI for their work without much oversight, sometimes even sharing sensitive company info with AI tools.
AI Use Goes Beyond Expectations
The report reveals that half of all executives would prefer managing with AI rather than humans. Yet, 34% aren’t even sure if they’re talking to a real person or an AI. That’s a big gap in understanding. Even more concerning, 38% of business leaders admit they don’t fully know what an AI agent is. Despite this, about a third of C-level leaders have shared proprietary data with AI to get tasks done.
The findings come from a survey of more than 1,000 U.S. workers. It shows a widespread trust in AI, even when it means bending or breaking rules. About 45% of employees trust AI more than their coworkers. Over half (52%) would use AI to make their jobs easier, even if it went against company policies. And 67% of top managers say they’d use AI even if it broke the rules.
Misuse in Regulated Industries
This pattern of AI misuse isn’t limited to startups or casual users. The report points out that even highly regulated industries are affected. For example, 60% of finance workers admit to breaking AI rules. About a third say they’ve used AI to access restricted data. In security and healthcare, similar issues are happening. Many security employees knowingly use AI in ways that aren’t approved, and over a quarter of healthcare workers would prefer to report to AI rather than a human boss.
Donnchadh Casey, CEO of CalypsoAI, explains that the internal risk is growing fast. He emphasizes that companies need to understand that employees at all levels are using AI without enough oversight. Many organizations are already seeing this happen, which could lead to serious security problems if not addressed properly.
Shadow AI: A New Kind of Shadow IT
Justin St-Maurice from Info-Tech Research Group calls this phenomenon “Shadow AI,” comparing it to Shadow IT — the use of unauthorized tech tools. Employees turn to AI because it helps them do their work faster and easier. The data shows that more than half of workers would still use AI even if their company said no. Some have already shared sensitive documents with AI tools, which raises major security concerns.
St-Maurice says the real issue isn’t necessarily disloyalty but that companies are falling behind in managing AI. Without proper controls, sensitive information can easily leak out. He suggests a balanced approach: companies should create a secure, approved AI “gateway” that logs activity, redacts sensitive info, and follows clear rules. Training employees on how to use AI responsibly is also key.
Casey agrees that blocking AI entirely isn’t the answer. Instead, companies should give employees access to AI tools in a controlled way. It’s about monitoring usage, setting policies, and having real-time controls. This way, businesses can benefit from AI without losing sight of what’s happening under the hood.
The survey was carried out in June by Censuswide, interviewing 1,002 full-time U.S. workers aged 25 to 65. The findings underscore the urgent need for organizations to rethink how they oversee AI use. As AI becomes more embedded in daily work, companies must find smarter ways to manage it — balancing productivity with security.
What do you think?
It is nice to know your opinion. Leave a comment.