A new survey by CommBox highlights that security concerns are a significant obstacle to the widespread adoption of AI in customer service, with unregulated usage becoming more common within companies.
The report highlights a cautious approach toward AI implementation, as 45% of customer service teams have integrated AI in some capacity, but just 15% of brands report extensive use of the technology. Security concerns top the list of barriers, cited by 32% of respondents, closely followed by uncertainties around implementation costs and internal resistance, both at 31%.
“As the buzz around AI has grown, we’ve seen rapid adoption and interest within the sector. We’re now coming towards the end of this initial wave, where issues such as security are now front of mind. This is particularly the case for larger enterprises, which have stricter guardrails in place when compared to small and medium-sized organisations, where employees can’t just download the latest AI tool without due diligence,” said Dvir Hoffman, CEO at CommBox.
One of the emerging issues is “shadow AI,” where employees use unsanctioned AI tools without IT oversight. As AI tools become more accessible, companies face new challenges in ensuring that sensitive information isn’t exposed or used to train AI models without proper governance. This lack of control could slow the sector’s progress unless addressed.
They are exploring its use for sentiment analysis (39%), personalized recommendations (37%), chatbot automation (37%), improving knowledge access (36%), and predicting customer needs (33%).