Indonesian Political, Business & Finance News

Impressive: Singapore's Foreign Minister Builds His Own AI Using NanoClaw and Raspberry Pi

| | Source: KOMPAS Translated from Indonesian | Technology
Impressive: Singapore's Foreign Minister Builds His Own AI Using NanoClaw and Raspberry Pi
Image: KOMPAS

Many world politicians are currently only discussing artificial intelligence (AI) regulations. However, Singapore’s Foreign Minister, Dr. Vivian Balakrishnan, has taken a markedly different step. Balakrishnan has openly showcased a personal AI assistant that he built himself to aid his daily work. He describes the smart assistant as a “second brain” for a diplomat. His custom AI can answer every technical question, research topics, draft speeches, and provide daily briefings. “This system has become invaluable; I wouldn’t even dare to turn it off!” Balakrishnan wrote, quoted by KompasTekno from his Facebook post. Balakrishnan is no ordinary politician merely following technology trends. He is a former ophthalmologist educated at Anglo-Chinese School and National Junior College, and a recipient of the President’s Scholarship for medical studies at the National University of Singapore in 1980. This virtual assistant system is built on two open-source foundations. The first is NanoClaw, an independent AI assistant based on the Claude model created by developer Gavriel Cohen. The system runs locally on a Raspberry Pi and connects directly to the minister’s messaging apps, such as WhatsApp, Telegram, Slack, and Discord. Second, Balakrishnan applies the “LLM Wiki” pattern proposed by former Tesla AI Director Andrej Karpathy. This pattern addresses the “amnesia” issue in AI that often forgets conversation context with each new session. For Balakrishnan, AI is a technology that must be built and used, not merely debated as abstract policy. “Diplomats who learn to work with AI will have a significant advantage, and I believe that advantage starts now,” Balakrishnan concluded. As a state official, data privacy is the top priority.

View JSON | Print