Hong Kong authorities’ recent move to integrate artificial intelligence (AI) into the 1823 complaint platform aims to reduce public frustration by improving case handling and governance efficiency, lawmakers and experts said. The system, which processes residents’ complaints, will use AI to better direct issues to the appropriate government departments, potentially shortening residents’ waiting times.

Lawmakers welcomed the AI enhancements announced on Thursday as a positive development. Duncan Chiu, a lawmaker representing the technology sector, highlighted the challenge of complaints being misdirected, especially in complex cases such as disputes involving fallen trees located near boundaries between public and private land. He noted that AI could assist phone operators by prompting more targeted questions, enabling clearer case triage and reducing the time residents spend navigating the system.

Chiu also addressed the delayed introduction of AI on the platform, attributing it to the complexities involved in developing technology capable of reliably processing Cantonese, which contains many regional accents and can mix with English. “There are many accents in Cantonese and may sometimes mix with English or other languages,” he explained. Training the AI to recognize colloquial references such as landmarks, for example the “golden flower in Wan Chai,” added to the challenge.

This summer, Chiu was part of a delegation of 67 lawmakers who visited Beijing to observe the 12345 Citizen Hotline Service Centre, where big data and AI are employed to dynamically route complaints to government departments with specific timelines for resolution.

Fellow lawmaker Johnny Ng Kit-chong suggested that employing AI more broadly in government could enhance service delivery beyond the 1823 platform. He proposed the technology might be used to monitor departmental responses to complaints, analyse visual evidence submitted by residents through computer vision, and even generate follow-up prompts to ensure complainants are satisfied with outcomes.

Michael Chau, a professor of innovation and information management at the University of Hong Kong, concurred that expanding AI use represented progress but emphasized that the system’s effectiveness had yet to be proven. He cautioned that increased AI autonomy raised risks, particularly concerning data privacy and potential errors. Chau stressed the importance of maintaining human oversight in managing the sensitive information processed and called for robust safeguards to prevent data breaches and unintended AI actions.

As Hong Kong moves forward with incorporating AI into public service platforms, balancing technological advancement with privacy protection and operational accuracy remains a key concern for policymakers and experts alike.