MAY 14 —More than 50 states have now endorsed the “Political Declaration on Responsible Military Use of Artificial Intelligence and Autonomy,” first rolled out in 2023 at The Hague.
The latter is synonymous with international humanitarian laws and crimes against humanity.
At one level, this may appear to be a modest diplomatic exercise. Yet at another, it represents one of the earliest attempts by the international community to prevent artificial intelligence from becoming the most destabilising military force since the invention of nuclear weapons. Why ?
The declaration emerged from a growing realisation that AI is no longer confined to laboratories, civilian applications, or commercial competition.
Rather, it is now embedded directly into military planning, intelligence gathering, logistics, cyber warfare, missile targeting, drone swarms, and battlefield decision-making.
The danger is not merely that machines are becoming more intelligent. The greater concern is that human beings may gradually surrender moral judgment to systems optimised only for speed, efficiency, and lethality.
This is why the declaration matters. The “Political Declaration on Responsible Military Use of Artificial Intelligence and Autonomy” seeks to establish norms before catastrophic precedents become entrenched.
It calls on states to ensure that military AI systems remain governed by international law, especially international humanitarian law.
It emphasises accountability, human oversight, reliability, testing, transparency, and safeguards against unintended escalation.
In essence, it is an attempt to preserve the role of human responsibility in warfare before autonomy begins to outpace ethics.
The urgency cannot be overstated.
Artificial intelligence is developing far faster than the diplomatic institutions meant to regulate it.
This was precisely the warning advanced decades ago by the American sociologist Robert K. Merton at Harvard University, who cautioned that science often runs ahead of moral deliberation.
Today, that warning appears prophetic.
Military establishments around the world are now engaged in an accelerating race to integrate AI into command-and-control systems.
The United States, China, Russia, Israel, the European Union, India, and several middle powers are all investing heavily in autonomous capabilities.
The fear is straightforward.
Once AI systems are connected to missile defence networks, autonomous drones, cyber retaliation systems, or nuclear command structures, the time available for human reflection shrinks dramatically. A battlefield decision that once took hours may soon occur in seconds.
This compression of decision-making time could make future wars more likely, not less.
In moments of crisis, AI systems may interpret signals incorrectly, misclassify intentions, or escalate responses beyond what political leaders originally intended.
Worse, rival powers may trust algorithms more than diplomacy itself.
The world is therefore entering an era where wars could begin not through deliberate political calculation, but through machine-speed misinterpretation.
This is especially dangerous in regions already burdened by unresolved geopolitical rivalries.
The Indo-Pacific is one such region. The South China Sea, the Taiwan Strait, the Korean Peninsula, the East China Sea, and the broader maritime corridors linking the Pacific and Indian Oceans are already crowded with military assets, surveillance systems, cyber operations, and increasingly autonomous technologies.
A miscalculation involving AI-assisted military systems in these areas could escalate rapidly before diplomats even have the opportunity to intervene.
For Asean, the implications are profound.
Southeast Asia has traditionally relied on strategic patience, dialogue, consensus-building, and preventive diplomacy.
Asean’s diplomatic culture was built on slowing tensions before they spiral into open confrontation.
Artificial intelligence threatens to undermine this culture if military systems become designed primarily for instantaneous reaction rather than political restraint.
The region cannot assume that AI governance is solely the responsibility of major powers.
Asean must develop its own principles on responsible military AI use, particularly through the Asean Regional Forum, the East Asia Summit, the Asean Defence Ministers’ Meeting Plus, and Track 1.5 and Track 2 dialogues involving academics, military officials, and technology experts.
Malaysia, as the former Chair of Asean in 2025 and current Coordinator of China-Asean Relations between 2025 and 2028, is particularly well positioned to encourage these discussions.
This does not mean Asean should oppose technological advancement.
On the contrary, AI possesses immense civilian benefits. It can improve healthcare, disaster management, education, maritime monitoring, food security, and energy efficiency.
Even military applications such as logistics, humanitarian assistance, search-and-rescue missions, and defensive monitoring can contribute positively to regional stability.
The issue is therefore not AI itself. The issue is whether humanity retains ethical control over the systems it creates.
History offers sobering lessons. Technological breakthroughs are often initially celebrated for their strategic advantages before their destructive consequences become fully understood.
Chemical weapons, nuclear weapons, biological experimentation, and cyber warfare all followed this trajectory.
Artificial intelligence may now be approaching a similar turning point.
What makes AI uniquely dangerous is that it operates at extraordinary speed while also possessing the capacity to learn, adapt, and interact across interconnected systems.
Unlike previous weapons, AI can potentially influence every domain simultaneously: land, sea, air, cyber, outer space, finance, communications, and public opinion.
The militarisation of AI could therefore reshape not only warfare, but the very psychology of international relations.
Trust may erode further between states. Suspicion may deepen. Pre-emptive doctrines may become normalised.
Governments may feel pressured to automate responses simply because rivals are doing the same.
This creates a classic security dilemma intensified by machine learning.
One state’s pursuit of technological security becomes another state’s perception of existential vulnerability.
The “Political Declaration on Responsible Military Use of AI and Autonomy” is therefore significant precisely because it attempts to slow this dangerous momentum before it becomes irreversible.
Critics will argue that declarations alone are insufficient.
They are correct.
The declaration is not legally binding. Major powers remain divided. Enforcement mechanisms remain weak. Verification is difficult. Strategic competition continues to intensify.
Yet norms matter.
International politics has always depended partly on the gradual construction of expectations, standards, and red lines.
Even imperfect norms can shape state behaviour over time.
The alternative — allowing military AI development to proceed without any ethical framework at all — would be far more dangerous.
Human civilisation now faces a defining question.
Can technological power remain subordinate to human wisdom ?
Or will humanity create systems whose speed, autonomy, and lethality eventually exceed political control itself ?
The answer may determine whether artificial intelligence becomes humanity’s greatest achievement — or its gravest strategic mistake.
* Phar Kim Beng is a professor of Asean Studies and director of the Institute of Internationalisation and Asean Studies, International Islamic University of Malaysia.
** This is the personal opinion of the writer or publication and does not necessarily represent the views of Malay Mail.
