Pachara pointed out that although Thailand must keep pace with the fiercely competitive global AI landscape, it raises the question of whether it is time for Thailand to enact laws to regulate AI. He suggested four approaches for AI oversight.
On 25 Sep 2026 GMT+7, Pachara Naripthaphan, a member of the Securities and Exchange Commission (SEC), expressed his views on the ongoing debate about AI regulation, asking whether Thailand should begin establishing legal frameworks on this issue.
Pachara noted that this question is especially important for Thailand at this moment, as it's not just about the safety or ethics of technology. Thailand is currently experiencing a large influx of investments in AI and data centers. Having or lacking clear regulatory frameworks directly affects how much benefit Thailand gains from these investments. He observed that the global regulation of AI faces three difficult challenges that Thailand cannot avoid.
The first issue is that technology advances much faster than laws. AI technology evolves monthly, whereas most existing regulations are based on concepts from the industrial investment era.
The second issue is the difficulty in defining regulatory criteria, because AI is used in vastly different contexts—from light uses like games to systems related to national infrastructure safety. It is impossible to apply a single standard universally; regulation must be risk-level based.
The third issue is the unclear identification of the responsible regulatory authority and what tools to use—whether strict licensing or flexible, case-by-case risk-based adjustments.
"I believe the most important lesson is to never apply industrial-era regulatory thinking to AI, as it will inevitably lag behind technology. What must be done is to design mechanisms agile enough to keep pace with rapid technological changes," Pachara said.
Another greater challenge is defining "AI" legally, which lacks international consensus. He cited the EU’s AI Act—the world’s first AI regulation—where legal scholars criticize its broad definition that nearly covers all software types, causing legal uncertainty for developers and users alike.
"The problem is if the definition is too broad, it risks over-regulating low-risk cases. If too narrow, high-risk systems might escape regulation. Thailand must carefully consider this before drafting laws, not just copy others wholesale," Pachara said.
According to research published in the Journal of World Trade (2024), major AI powers adopt very different approaches. The US favors market-driven innovation first, then regulation. The EU implements comprehensive AI laws covering the entire AI lifecycle. China uses a mixed approach promoting market capacity alongside data security controls. Singapore emphasizes voluntary frameworks guiding organizations to design and use AI responsibly, backed by proactive digital trade agreements.
"Singapore is probably the model closest to Thailand’s context, as a smaller country leveraging agility and practical standards rather than competing with Europe’s resource-intensive strict regulations," Pachara said.
Pachara also stressed that Thailand’s role is primarily as a technology user. Thus, regulatory frameworks must be designed from the user perspective to protect the public in a timely manner, with broad but precise focus. Especially as frontier and agentic AI increasingly make autonomous decisions and actions, considerations must include social impacts like public trust and dependence, legal issues like accountability for AI errors, industrial impacts such as workforce and business adaptation, and economic aspects like fair value distribution within Thailand’s economy, before deciding how to design mechanisms carefully without rushing.
Pachara summarized that although strict regulation is not yet necessary, relevant agencies should comprehensively address arising issues. He proposed four approaches:
1. Adopt risk-based regulation, not uniform controls, to avoid burdening small businesses and early-stage innovations, while imposing stricter standards on AI affecting public safety or fundamental rights.
2. Use existing laws to tackle AI-related problems immediately, such as voice fraud or employment discrimination, applying current consumer protection and labor laws without waiting for new legislation.
3. Promote Thailand’s AI stance in international digital trade negotiations to avoid missing out on global standard-setting and protect Thai business interests in world markets.
4. Create adaptable regulatory mechanisms rather than traditional bureaucratic agencies, with periodic review and adjustment of rules alongside private sector, academics, and civil society, to keep pace with technological changes.
Pachara added that beyond safety and ethics, another dimension is preserving economic value from incoming AI investments. Flexible regulation should not only cover safety and ethics but also link investment incentives to domestic value creation, such as tying BOI privileges to value addition, workforce training, integration with Thailand’s existing strong electronics industry base, and clear early frameworks on energy and water use in data centers—since later fixes are costlier than upfront design.
"I believe Thailand does not need to rush into strict AI laws like Europe now, but also should not remain directionless. The right balance is to start setting flexible principles, maximize use of existing laws, and tie investment conditions to real domestic value creation... think broadly, focus precisely, but without haste," Pachara concluded.