Complex integration process, data privacy hinder AI chatbot adoption: Report

Perceptions about chatbots generating ‘cold and static’ responses hold back many from adopting artificial intelligence-based chatbots. As per a study by customer support automation platform Kaputure CX, 50% of the surveyed customer support managers from Indian B2C brands said that this is a major factor hindering them from adopting AI chatbots.

The study attributes this hesitancy to confusion between rule-based chatbots and AI-powered ones. Unlike AI chatbots which are more dynamic, rule-based chatbots operate on predefined conversational paths and offer responses to predetermined questions and answers.

Further, 19% of the respondents said that complex setup and integration process is a major obstacle in AI chatbot adoption. The complexity of chatbot increases with its sophistication levels. To develop an effective AI chatbot, obtaining and processing relevant data, along with fine-tuning the underlying model can also prove to be deterrents. Lastly, enterprises handling personal and critical data would have to adhere to various regulatory frameworks such as GDPR and HIPAA.

Concerns on data privacy also causes apprehension, said 17% of the respondents. This is particularly pronounced in industries like banking, financial services, and insurance (BFSI) and healthcare.

Lastly, a lot of customers would value real-time conversations with a human executive rather than with a bot. Even against benefits like potential for increased efficiency and cost saving, 14% of the managers said that human touch in customer interactions is essential for building relationships, and automation might lead to a loss of empathy and emotional connection. 

A June report by McKinsey on the economic impact of generative AI said that the technology could add trillions of dollars in value to global economy. Across 63 use cases analysed by the company, generative AI can add $2.6 trillion to $4.4 trillion annually. This estimate would approximately double if the impact of embedding generative AI into software that is currently used for other tasks beyond those use cases, is also included, the report added.