Expanding the human touch with generative AI

For the past decades, the ideology propelled by New Public Management suggested that government performance would improve through a more business-like approach focusing on efficiency. While this approach largely worked well, it fell short for a segment of citizens, sometimes with severe consequences.

The report by the Temporary Committee on Implementation Organizations (TCU) in 2021 revealed that as much as 20 percent of citizens couldn't adequately understand laws and regulations across multiple policy areas. The call for more human-centered approaches in governmental actions has therefore been a focal point for several years, further amplified by incidents like the childcare allowance affair, the messages in Wijdemeren, and the handling of earthquake damage in Groningen. However, government spending continues to rise, risking further pressure on the human aspect.

Smart utilization

In recent years, there has been a greater emphasis on providing personalized services to citizens. This is beneficial as it can better address the needs of a significant portion of citizens. However, it often leads to additional costs. Meanwhile, governmental spending has surged in recent years and is likely to continue rising in the future. Therefore, the government has been advised to cut €17 billion in spending by top civil servants of the Budgetary Space Study Group.

Potential future limitations on government spending carry the risk of further compromising the human aspect and shifting the focus to efficiency, possibly through increased standardization. Personalized services thus risk fading into the background. To prevent this, it's crucial to devise an innovative solution that promotes both efficiency and personalized, accessible service.

This calls for smart utilization of resources to efficiently provide personalized services with a human touch. Generative AI has the potential to play a key role in addressing this issue.

What is Generative AI?

Generative AI is a form of artificial intelligence that utilizes advanced algorithms and machine learning techniques to create new content. This content can range from text and images to source code, music, and even videos. The process starts with a prompt – a question, topic, or task – presented by the user to the AI. Based on this prompt, the AI analyzes the request using Large Language Models (LLMs), which are capable of processing and interpreting large amounts of data.

In recent years, generative AI has undergone remarkable development, greatly expanding the possibilities for governmental services.

In response to this development, the government presented a vision in January regarding their perspective on the use of Generative AI. In addition to risks and challenges, the government also sees opportunities regarding productivity growth and new tasks and responsibilities within organizations and the economy as a whole. Specifically for government services, we also see opportunities for the use of Generative AI. It can even play a key role in preserving the human aspect at a time when government spending is under pressure. For this, we identify four ways:

1. Making services more accessible

It's often complicated for citizens to know which government agency to approach. Generative AI can help by creating a single digital portal for various government agencies and addressing citizens' most pressing questions.

Example: virtual assistants for citizen assistance

An AI-driven assistant on the government website answering questions such as: where can I apply for a new passport, how do I apply for child allowance, or what are the requirements for housing benefits?

Example: virtual assistants for form filling

An AI assistant guiding the citizen through form filling, such as: help me fill out the tax refund form. The virtual assistant then provides detailed instructions, including required documents and steps.

2. Reducing the complexity of laws and regulations

With constantly changing and increasingly complex laws and regulations, the public sector faces the challenge of keeping them executable for employees and understandable for citizens. Generative AI can help by analysing large amounts of legislation, making them understandable in plain language, and then applying them to specific situations.

This makes it much easier for employees to get the right information, especially in the case of complex, infrequent, and/or multi-law issues. Because these are impactful decisions, the employee will always make the assessments themselves. Thus, Generative AI is nothing more than a tool.

Example: automatic analysis and summarization of legislation

Loading various laws into an LLM to summarize complex legal documents. This summary can be tailored to a specific situation. For example, you can request: Summarize the regulations for disposable plastics at a B2 level, specifically for pickup locations.

3. More efficient process execution while maintaining the human aspect

How can processes be designed to be faster and more efficient without compromising the human aspect? Generative AI can support this by automating administrative tasks, allowing employees to focus on more complex tasks that require personal attention to citizens.

Example: automation of routine administration

Setting up a system to automate the processing of standard permit applications. The system is given criteria so that these applications can be automatically identified.

4. Accelerating the policy cycle with an additional focus on the human aspect

Finally, Generative AI can also play a very valuable role in policy development by simulating various policy scenarios, analyzing large amounts of data and external trends, formulating ideas and suggestions for societal issues, and drafting policy briefs.

While Generative AI cannot replace policy development, it is a supportive tool for policy officials to accelerate and enrich the policy formation process. Moreover, Generative AI enables assisting in assessing whether the formulated policy adequately considers the human aspect.

Example: policy scenario simulation

Generative AI can simulate the impact of policy changes, such as: Simulate the impact of raising the minimum age for alcohol consumption. The AI application then conducts a simulation on, for example, the social and economic consequences.

Example: data analysis for policy development

Generative AI is capable of analysing large amounts of data to map trends for policy development. An example question could be: Analyse public transportation data and urban mobility trends to make recommendations for the 2025 urban transportation plan.

Conclusion

Generative AI has the potential to play a crucial role in increasing the human aspect in government actions, a growing need recently emphasized by incidents such as the childcare allowance affair. While rising government spending poses a challenge, Generative AI offers opportunities to maintain personalized services, with costs eventually being reduced.

Read the original article in Dutch