AI in public service delivery: What Bangladesh needs to consider
As the use of AI in public service delivery increases worldwide, a balanced approach should be taken into account to effectively implement AI’s potential applications in Bangladesh
Bangladesh is on a significant developmental trajectory with its large population and fast-paced economic growth. To sustain this progress, the effective delivery of public services is crucial.
In this context, Artificial Intelligence presents a unique opportunity and understanding AI's current and potential applications in Bangladesh's public sector is essential now.
More sophisticated models of AL can bring greater efficiency and personalization while emerging technologies like the Internet of Things (IoT) and blockchain could further revolutionise these services.
However, to fully grasp the impact, it is important to first look at the current applications of AI in Bangladesh. Though still in its early stages, AI is already being used in several key sectors.
Keep updated, follow The Business Standard's Google news channel
AI-driven diagnostic tools and telemedicine platforms are gaining momentum while the 'e-Agriculture' platform is helping farmers access timely, relevant information. Additionally, AI plays a vital role in predicting natural disasters, particularly through early warning systems that mitigate the effects of floods and cyclones.
AI and the digital divide
Despite its potential, integrating AI into public services in developing countries like Bangladesh has significant challenges. One of the most pressing issues is the digital divide. While access to the internet and digital devices has improved, a large portion of the population—particularly in rural areas, among the elderly, and within low-income groups—still struggles to benefit from AI-driven services.
Similar challenges have been seen in Ethiopia, where AI reliance on public services has exacerbated the digital divide, and in Nepal's education system, where students in remote areas are unable to fully benefit from online learning tools due to poor internet connectivity.
The fear of job loss
AI can create significant job losses in the public sector, particularly in Bangladesh, where unemployment is already a concern. Integrating AI systems into administrative and operational roles could exacerbate the situation.
For example, in the Philippines, the automation of certain public services, such as government office tasks, has displaced many low-skilled workers. Similarly, in Mexico, AI in the manufacturing sector resulted in widespread job losses among factory workers.
The 'data security' issue
The integration of AI in public services requires extensive data collection and analysis, but Bangladesh has yet to develop robust data protection regulations. Without flawless data security, personal information becomes vulnerable to misuse and cyberattacks.
Kenya's experience with the Huduma Namba biometric ID system offers a relevant lesson, where AI has sparked significant privacy concerns. Likewise, Brazil's deployment of AI in policing and surveillance, without strong data protection, has led to mass surveillance and potential data misuse.
The ails of AI: Reinforcing societal biases and the 'AI algorithms' problem
Moreover, AI systems can unintentionally reinforce societal biases. In Bangladesh, where biases related to gender, class and ethnicity are prevalent, AI algorithms trained on biassed data can result in discriminatory practices.
For instance, India's Aadhaar-based Public Distribution System (PDS) sometimes fails to recognise biometric data, denying rightful recipients access to food rations. Similarly, in South Africa, AI-driven credit scoring systems, often based on biassed data, have led to unfair lending practices that disproportionately affect marginalised communities.
Successful AI implementation requires both robust infrastructure and skilled human resources. While Bangladesh has made strides in the IT sector, challenges such as insufficient internet bandwidth and a shortage of trained AI professionals remain obstacles.
In public service delivery, where decisions can significantly impact individuals' lives, the lack of transparency in AI algorithms can pose severe ethical dilemmas. For instance, if an AI system wrongly denies a citizen access to social benefits or health services, determining accountability can be difficult, leaving affected individuals without alternatives.
In some parts of Nigeria, AI systems proved to be problematic in sectors like healthcare where AI-driven diagnostic tools are used without clear guidelines or accountability mechanisms.
AI systems lack empathy that human service providers can bring to public service delivery. In contexts such as healthcare and social services, the human touch is crucial.
For example, automated responses in social welfare programs might not be able to fully realise the complexities of individual cases, leading to inadequate support for those in need.
Not all is bleak
There are some success stories from which Bangladesh can take lessons.
For instance, Rwanda, a developing country in East Africa, has made substantial strides in leveraging technology to improve public service delivery. In 2017, Rwanda in collaboration with several technology companies, including Babylon Health, to launch an AI-powered digital health service called 'Babyl.'
The service integrates AI with mobile phone technology to provide accessible healthcare to Rwandans, particularly those in remote and underserved areas. Babyl has extended health care access to rural and remote areas where health care facilities and professionals are scarce.
With a high mobile phone penetration rate in Rwanda, many people can access healthcare services without travelling long distances.
The AI system helps streamline the healthcare process by providing initial assessments and reducing the burden on human doctors. This efficiency allows healthcare providers to focus on more critical cases and improves overall service delivery. The success of Babyl in Rwanda proves the potential for scalable AI healthcare solutions in other developing countries with similar challenges.
An objective and well-structured policy ecosystem is essential to fully harness the potential of AI in public service delivery. Bangladesh must prioritise investments in digital infrastructure to ensure reliable internet connectivity, especially in remote areas.
Additionally, developing a skilled workforce capable of managing AI systems is critical. Comprehensive regulations surrounding data privacy, security and the ethical use of AI are equally necessary. Ethical guidelines and standards should be developed.
Integrating data from diverse sources is also key to creating comprehensive datasets that AI systems can use for accurate analysis and decision-making. Collaboration between industry and academia will be vital to bridging the gap between theory and practice and staying up-to-date with the latest AI advancements.
Additionally, public-private partnerships (PPP) are crucial for fostering innovation and resource-sharing in AI technology development. Partnerships with developed nations, both at the state and non-state level, can facilitate technology transfer and capacity building. Moreover, engaging with non-resident Bangladeshis (NRBs) can open avenues for knowledge-sharing and investment, further accelerating AI-driven progress.
While AI holds immense promise for transforming public service delivery, the potential risks cannot be ignored. Therefore, a balanced approach, taking into account both technological advancements and socio-economic factors, will be crucial for effectively integrating AI into Bangladesh's public services.
Dr Mohammad Kamrul Hasan is a public administration researcher and practitioner. mkhmiraz@gmail.com.
Disclaimer: The views and opinions expressed in this article are those of the author and do not necessarily reflect the opinions and views of The Business Standard.
