Most Pakistani professionals using AI tools without formal training: survey

LAHORE: A majority of professionals in Pakistan have begun using artificial intelligence (AI) tools at work without receiving any formal training, while most organisations have also failed to provide structured learning opportunities, a new survey has revealed.

The research, titled ;Cybersecurity in the Workplace: Employee Knowledge and Behaviour’ and conducted by Kaspersky, found that 86 per cent of professionals surveyed in Pakistan use AI tools for work-related tasks. However, only 52 per cent reported receiving any training on the cybersecurity aspects of using neural networks, an essential safeguard against AI-related risks such as data leaks and prompt injections.

Interestingly, many employees across various organisations said they use AI tools to write and edit text, particularly for work emails. An overwhelming 98 per cent of respondents said they understand the term ‘generative artificial intelligence (Gen AI)’, and for many, this knowledge has moved beyond theory — AI tools have now become part of their daily workflows.

Of those surveyed, 86 per cent said they use AI tools for work, with 68 per cent using them to write or edit text, 52 per cent for work emails, 56.5 percent to create images or videos using neural networks, and 35 per cent for data analytics.

The study also highlighted a serious gap in employee preparedness for AI risks. About 21 per cent of professionals said they had received no AI-related training at all. Among those who had, 66 per cent said the focus was on how to use AI tools effectively and create prompts, while just 52 per cent received guidance on cybersecurity practices.

Although AI tools are increasingly automating everyday tasks and becoming commonplace in many organisations, they often fall under ‘shadow IT’ — tools used by employees without official approval or oversight. According to the survey, 81 per cent of respondents said Gen AI tools were permitted at their workplace, 15 per cent said they were not allowed, while 4 percent were unsure about their organisation’s policy.

Cybersecurity expert Arslan Bukhari said that to make employee use of AI clearer and safer, organisations should adopt a well-documented, company-wide policy. “This policy should prohibit AI use in specific functions or with sensitive data, define which AI tools are authorised for use, and ensure that only approved applications are accessible to employees,” he explained. He added that firms must develop comprehensive frameworks to address all relevant risks associated with AI use.

To strengthen corporate AI security, the Kaspersky report recommended that organisations: train employees on responsible AI use; equip IT specialists with knowledge of exploitation techniques and defence strategies; and ensure cybersecurity solutions are installed on all devices used to access business data, whether work or personal.