Despite significant technological progress, the quality and efficiency of public services are still lacking across the world. Citizens expect them to be as user-friendly and fast as banking services or taxi apps, but governments have been struggling to keep up. How can this be changed? By using AI.
Officials increasingly recognize the potential of AI for their public services. According to the 2023 AI Index report from Stanford University’s Institute for Human-Centered Artificial Intelligence, government spending on AI surged to over $3.3 billion in fiscal year 2022, marking an increase of nearly $600 million from the previous year. Moreover, a Deloitte report shows that governments are expected to launch more pilot projects to explore generative AI's capabilities and ensure its safe, secure application.
Let’s look at some interesting ways in which this emerging technology can be leveraged in the public sector.
Public Inquiries
One of the early pioneers in introducing AI-powered tools into government services was Singapore back in 2014. They created a virtual assistant that provided residents with information on various topics such as tax filings, passport renewals, and social service applications. The assistant used a natural language processing engine to interpret public inquiries accurately and generate appropriate responses.
Thanks to this capability, it could ask additional questions to narrow down the initial response, ensuring it met the specific needs of the user. To date, it has been deployed on 70 government platforms and has proved to increase satisfaction among citizens.
Immigration Processes
Another example is an AI chatbot created in 2015 by the U.S. Citizenship and Immigration Services (USCIS). It answered a wide range of questions about immigration procedures, including visa applications, green cards, and naturalization processes. The implementation of the bot has successfully reduced waiting times for thousands of users and lessened the load on human customer service agents at USCIS.
It was reported that the bot was able to engage with more than 3 million users annually, providing quick answers to routine questions.
Disease Prediction
AI tools can significantly benefit the public health sector as well. For instance, a Canadian AI-driven health monitoring platform gained international recognition for identifying the outbreak of COVID-19 before the World Health Organization released its warning.
The platform uses big data analytics to track, locate, and conceptualize the spread of diseases. Its early warning helped various governments and healthcare organizations prepare and respond more effectively to the pandemic.
Public Engagement
AI can also enhance public engagement in the decision-making process. While a team of even dozens of employees may struggle to thoroughly read and analyze all citizen suggestions, AI can efficiently manage this task. Furthermore, it can classify ideas, summarize trends, and group similar contributions by different criteria such as theme, demographic traits, or location.
For example, in early 2019, a rising number of Belgian young people were protesting against climate change inaction. In response, an AI-based participation platform was launched, allowing users to submit their ideas for tackling climate change. Over three months, users submitted 1,700 ideas, 2,600 comments, and 32,000 votes on initiatives they supported.
The AI system then analyzed these materials, identified and grouped the most significant and popular priorities, and developed a report for officials containing 16 policy recommendations.
Big Challenges
There is no question that AI is revolutionizing industries, including public services. However, significant challenges make the mass integration of this technology tricky. The most pressing one, in my opinion, is privacy. AI often relies on large datasets that may include sensitive information from individuals who did not consent to their data being used.
Therefore, moving forward, we must ensure that this technology not only improves the efficiency of public services but also uses data in a secure and ethical way. How to ensure that is an open question.