Employers banking on their policy to control AI use are in for a reality check
More than half of employees in the United States will risk violating company policy to use artificial intelligence tools to make their job easier, according to a new report.
Findings from CalypsoAI's Insider AI Threat Report revealed 52% of employees admitted they would use use AI to ease workloads, even if it breached policy and a quarter have already used AI without checking if it's allowed.
More than a quarter also confessed they submitted proprietary company info so AI could complete a task (28%) and used AI to access sensitive data or documents (28%).
"Employees increasingly see AI as a tool for efficiency or leverage, not a shared risk," the report read.
The latest insight into the secretive use of AI follows a recent Cox Business survey of more than 1,000 Gen Z and Millennial employees in the US that found younger employees were hesitant to admit they use the technology because of fear of losing work.
Disconnect in values
Reasons for the unauthorised use could be vary from blind faith in leadership and confusion about company policies.
The report found that 84% of employees are somewhat or very confident that their company's CEO or IT team would be able to catch an AI-led breach, while frontline employees, 21% said their company's AI policies were unclear, so they use AI as they wish.
Being disconnected from the company's values may also be a factor, as 30% of frontline staff said they don't care about their company's AI policy and 37% said they won't feel guilty if they broke AI protocol.
If they couldn't use AI, 25% said they would quit their job.
Donnchadh Casey, CEO of CalypsoAI, warned that frontline employees' unsupervised use of AI should be a "wake-up call" for organisations.
"We know the inappropriate use of AI can be catastrophic for enterprises, and this isn't a future threat - it's already happening inside organisations today," Casey said in a statement.
Lack of C-suite knowledge
The findings come in the wake of widespread adoption of AI tools, including AI agents, in workplaces.
Previous findings from Workday revealed that 82% of organisations are deploying AI agents to reduce employee workloads and set the company up for financial success.
But CalypsoAI's poll found that 38% of leaders have no idea what an AI agent is, with 10% not being able to tell the difference between an AI agent and a virtual employee.
"Leaders are green-lighting AI faster than they can explain it," Casey said in the report. "You can't govern what you don't understand, and that disconnect is now a top-tier enterprise risk."
According to the CEO, uncritical enthusiasm for technology can quickly unravel progress for organisations.
"Without the right awareness, blind eagerness will have detrimental impacts for enterprises when it comes to AI implementation," he said.
"Organisations need both technical controls and compliance, along with tools to educate employees and create a broader culture of respect for the power of AI."
