More than two-thirds of workers say their employers do not adequately support the use of artificial intelligence at work, according to a new study. The result is a growing wave of employees turning to unauthorized AI tools, even when they understand the security and compliance risks.
Researchers found that the primary barrier is not a shortage of technology. The real problem is a mismatch between available AI tools and the daily realities of how people actually work. Many organizations have not updated policies or training to match the rapid adoption of generative AI.
The Gap Between Policy and Practice
The study surveyed thousands of office workers across industries. It found that employees often bypass official channels because approved AI systems are too restrictive or do not fit their specific tasks. Instead, workers rely on consumer-grade AI platforms for tasks like writing, data analysis and scheduling.
This behavior creates security vulnerabilities. Company data may be exposed to third-party servers. Intellectual property risks increase. Yet workers feel justified because their productivity depends on these tools.
Why This Matters
The findings highlight a growing disconnect between corporate IT policy and real-world work habits. For employers, the cost is not just security breaches but also lost visibility into how work gets done. For employees, the risk includes accidental data leaks and potential disciplinary action.
Organizations that fail to adapt may see a rise in shadow AI, a term for the unauthorized use of AI tools. This trend mirrors the early days of cloud computing, when employees adopted Dropbox and Google Docs without official approval.
The study suggests that companies should focus less on blocking tools and more on aligning AI offerings with the actual workflows of their teams. That means investing in flexible, secure AI platforms that workers actually want to use.
Until then, the gap between policy and practice is likely to widen.
