On paper, everyone would agree that AI makes innovation projects in the Civil Servant Candidate (“CPNS”) Basic Training Program (Pelatihan Dasar/Latsar) much easier. However, what is perhaps more interesting is not simply whether AI simplifies these projects, but how it transforms the challenges that program participants face. The author outlines six projections regarding challenges that may become increasingly relevant to innovation projects carried out by CPNS Basic Training participants within the Supreme Court of the Republic of Indonesia.
This article stems from the author’s reflections while participating in the CPNS Basic Training, where an innovation project, a web-based application, was developed with significant assistance from AI. This experience continued as the author developed several other applications for their assigned workplace alongside other court personnel. Therefore, the views expressed represent the author’s personal opinions shared to convey these practical experiences. The author acknowledges that they are not an expert and lack formal education in computer science; consequently, this article is shared with the understanding that it contains limitations.
1. The Growing Importance of Data Protection
The judicial system manages a wide range of sensitive information, ranging from personal data of parties to confidential case documents.
In the context of the Supreme Court’s CPNS Basic Training innovation project, the ability to utilize AI must go hand in hand with the ability to maintain data confidentiality. Participants will likely need to understand which information is appropriate to process using public AI models and which information must remain within the agency’s internal environment. Moving forward, topics such as masking techniques and local encryption will likely become essential knowledge for training participants.
For example, using locally deployed AI allows sensitive data to stay within the agency’s secure network. Conversely, when using API-based AI services, robust security mechanisms must be applied, such as masking personal data before transmission or separating local AI for sensitive data from API-based AI for non-confidential data.
2. Prompt Engineering as a New Core Competency
Some may assume that AI only requires simple questions to produce great results. In reality, the quality of AI output is heavily influenced by the precision and structure of the instructions provided. The author describes this as a “cat and mouse game with AI”, simply put, users must be able to debate, guide, or correct the AI.
This is where basic technical and coding literacy becomes crucial. For example, the author has occasionally had to correct the AI by pointing out flaws in its logic, prompting the AI to admit its oversight and adjust its response. It is reasonable to expect an increasing number of training sessions and workshops on Prompt Engineering as part of institutional competency development in the future.
3. Investing in AI Credits and Tokens as a Measure of Productivity
This challenge is relatively straightforward. Most AI services operate on subscription plans and credit or token systems. As a result, productivity is no longer influenced solely by individual competence or the budget to hire a professional developer, but also by the volume of AI credits or tokens a user commands. Put simply, the capital required for modern creative and technical work now frequently takes the form of AI tokens.
4. UI/UX Intuition: A Distinctive Edge That Remains Hard to Replace
In the author’s experience, instructing AI to build a website or application without explicit design parameters rarely yields a modern or intuitive user interface. Mainstream AI models tend to produce functional but default layouts, requiring users to explicitly prompt for modern UI standards.
A foundational understanding of color theory, user experience (UX), and digital design trends serves as a vital asset when guiding AI. For instance, AI does not always automatically deduce basic UX conventions, such as placing a prominent “Back to Dashboard” button, establishing clear visual hierarchy, or visually distinguishing major call to action buttons from secondary ones.
5. Stakeholder Engagement as the Ultimate Differentiator
If AI enables rapid application development, the primary metric for a successful project is no longer whether an application functions, but whether it is sustainable and impactful. Crucial questions now arise: Is the innovation actively used by people within the work environment? To what extent does it streamline operational processes? Does it provide tangible benefits to the public seeking justice?
Stakeholder engagement extends well beyond post launch maintenance into the early development stages. The ability to conduct outreach, secure buy in, involve interdepartmental units (for example, between the clerk’s office and the secretariat), and gather user feedback during trial runs will carry significant weight.
AI dramatically accelerates the build phase, shifting the developer’s time allocation toward human centric tasks. For example, in a 30 day timeline, building a complex application might have previously taken 14 days, leaving 16 days for dissemination. With AI, that same application might be built in a single day, allowing (and requiring) developers to spend the remaining 29 days gathering feedback, conducting user stress tests, and driving adoption.
6. Hardware Integration as a Value Multiplier
The evolution of AI is shifting the added value of digital innovations from pure software development toward hardware integration. With AI assistance, building web applications on standard laptops or workplace servers has become accessible. However, future high value innovations will likely be defined by their ability to interface software with physical hardware, such as cameras, sensors, QR scanners, and biometric devices.
Discussions around hardware specifications will become far more commonplace among non technical staff. Questions like “Which GPU equipped server configuration is required to run this model locally?” will enter standard project planning. Ultimately, the future of civil service innovation will depend on two pillars: broad stakeholder engagement on the human side, and deep hardware integration on the technical side.
Conclusion
These six projections are by no means exhaustive. Rather, they reflect the author’s personal observations on how artificial intelligence is reshaping innovation projects within the Supreme Court’s CPNS Basic Training Program.
As a junior civil servant at the beginning of this professional journey, the author offers these reflections not as definitive scientific conclusions, but as a practical contribution to the ongoing conversation surrounding AI integration in public sector training.
Untuk Mendapatkan Berita Terbaru Suara BSDK, Follow Channel WhatsApp: SUARABSDKMARI