UI academic urges Indonesia to adopt risk-based AI regulation
The European Union Artificial Intelligence Act (EU AI Act) can serve as a reference for Indonesia in designing risk-based AI regulation, but that does not mean the model must be applied identically.
Purwokerto (ANTARA) - Parulian Paidi Aritonang, an academic from the University of Indonesia, has urged Indonesia to strengthen risk-based artificial intelligence (AI) regulation by taking global regulatory developments as one of its references.
“The European Union Artificial Intelligence Act (EU AI Act) can serve as a reference for Indonesia in designing risk-based AI regulation, but that does not mean the model must be applied identically,” said Parulian at the 4th International Conference on Law, Governance, and Social Justice (ICOLGAS) 2026 in Purwokerto, Banyumas Regency, Central Java, on Tuesday.
The Dean of the UI Faculty of Law explained that a risk-based approach is needed to differentiate AI use according to the level of risk involved, ranging from prohibited practices, high-risk systems, and transparency obligations, to low-risk uses.
According to him, the regulatory framework must also govern obligations for high-risk AI systems, including risk management, data governance, technical documentation, record-keeping, human oversight, accuracy, system resilience, and cybersecurity.
“What needs attention is not merely the technology, but how emerging risks can be identified, controlled, and monitored through a clear legal framework,” he said.
In addition, he said there are specific obligations for general-purpose AI (GPAI), including technical documentation, copyright policies, and summaries of the materials used in training.
Meanwhile, Indonesia to date has no single comprehensive law specifically governing AI in the way the EU AI Act does.
According to Parulian, a number of existing regulations could serve as entry points for AI governance, including provisions on electronic systems and personal data processing.
“Indonesia needs to assess whether the existing sectoral approach is sufficient or whether it should move towards risk-based AI governance,” he said.
Parulian also highlighted the importance of governance and law enforcement, encompassing oversight, conformity assessments, registration, post-market monitoring of systems, incident reporting, and the imposition of sanctions.
“That way, strengthening AI regulation is directed not only at encouraging innovation, but also at ensuring that the technology is used responsibly through adequate oversight and protection mechanisms,” he said.