Editor’s Note
by Siu Tzyy Wei, Lead Editor - Maritime Crescent Desk
Power is easiest to understand when you can see who holds it.
This week, Brunei strengthens the capabilities that could see it take on a greater role in regional disaster relief. Indonesia, meanwhile, moves closer to giving the state a power anti-corruption reformers have wanted for decades: the ability to seize assets without first securing a conviction. Malaysia, meanwhile, finds itself at the centre of another contest over power: whether attracting the infrastructure behind the AI race is enough to give it a stake in the technology itself.
Three different kinds of power, three very different questions about what happens when it is put to use.
Because the question is never just who has the power. It is who gets to hold it, and what keeps them from holding it too tightly.
Brunei Darussalam 🇧🇳
Brunei, PACAF, and Interoperability in the Right Domain
by Wira Gregory Ejau, in Bandar Seri Begawan
From August 25 to September 5, US Pacific Air Forces and the Royal Brunei Air Force conducted PACIFIC ANGEL 26-1, a joint search and rescue exercise staged at Rimba Air Base in Bandar Seri Begawan. The exercise covered overland and overwater personnel and equipment airdrops, aeromedical operations, cargo handling, and humanitarian assistance and disaster relief response.
While Brunei’s armed forces are small, well-funded relative to the state’s size, and professionally capable, its forces are not necessarily configured for high-intensity combat roles alongside a major power. In the development of its operational toolkit for regional emergency contingencies, PACIFIC ANGEL develops durable and arguably more immediately useful interoperability in the domain where Brunei’s military can realistically contribute to regional operations. Southeast Asia is among the world’s most disaster-prone regions, with its exposure to typhoons, earthquakes, and flooding across a maritime geography where rapid aerial response is not an enhancement but a necessity. A Royal Brunei Air Force capable of integrating seamlessly into PACAF-led HA/DR operations is a force multiplier that extends well beyond what Brunei’s own assets could achieve independently.
PACIFIC ANGEL follows Pahlawan Warrior 26, which officially concluded on the 20th of August at Tutong Camp, where a US Army and Oklahoma National Guard exercise with the Royal Brunei Land Forces covered jungle warfare, land navigation, medical evacuation, and unmanned aircraft systems. Before that, US Pacific Air Forces conducted air domain awareness training with the Royal Brunei Armed Forces in January, and a US Air Force MQ-9 Reaper was deployed to Brunei in May. The cumulative picture across 2026 of bilateral military relationship develops a consistent pattern and depth across air, land, and SAR domains simultaneously.
For a small state, interoperability with the dominant regional military power is a form of presence within a security architecture that the smaller state would be unable to sustain through its own resources alone. This embeds the Royal Brunei Armed Forces more firmly into the potential for future operational and planning frameworks of US Pacific Command, establishing Brunei as a reliable and integrated partner in its regional posture.
What the concentrated exercise calendar of 2026 ultimately suggests is that the Brunei-US military relationship is deepening, and that the domain in which Brunei is investing most visibly is precisely the one where its contribution to regional security is most credible and most sustainable.
Gregory is an MSc candidate in Strategic Studies at the S. Rajaratnam School of International Studies (RSIS), Nanyang Technological University. He works as a freelance writer specializing in international history, conflict, and counterterrorism, with experience in academia, investigative journalism, and voluntary uniformed service. He currently provides research assistance with the International Institute for Strategic Studies (IISS) under their Southeast Asian Security and Defence Internship Programme and conducts investigations on regional security and transnational crime for a confidential company.
Indonesia 🇮🇩
Careful What You Wish For
by Rayhan Prabu Kusumo, in Jakarta
On August 27, a crowd gathered outside the parliament in Jakarta. Many had come from Pati, a regency five three hundred kilometres away. By the end of the afternoon they succeeded in pressuring leaders to sign a letter promising to pass the Asset Forfeiture Bill by the fifteenth of December, and pledging to resign if they failed. A street protest that ends in a signed promise and a resignation pledge is not how these things usually go here.
This demand is old and, on the surface, entirely reasonable. For decades, anti-corruption reformers have asked for a law that lets the state take back stolen wealth, and for just as long, governments have found reasons not to pass it. The bill would let prosecutors and law enforcement seize assets tied to a crime, in some cases without waiting for the owner to be convicted. Countries serious about corruption have tools like this. Indonesia, which claws back only a fraction of what is stolen from it, plainly needs one.
The trouble is that the feature which makes the law powerful is the one that makes it dangerous. Seizing a person’s property before proving them guilty reverses the order a legal system usually runs in. The burden falls on the owner to show the money is clean, and the constitution’s protection of property and of innocence begins to bend.
Take Tom Lembong, prosecuted in a sugar-import case and then pardoned by presidential decree, an enforcement machine switched on and off by politics. Now picture that machine able to freeze your house while it decides. Members of the House writing the bill admit fearing it themselves, that it could become a tool to attack opponents. Others argue that money laundering may be quietly kept out of its reach, which would leave it fearsome to the weak and toothless against the strong.
Now, as this brief once probed, Indonesia’s anti-corruption enforcement is split among rival forces that investigate one another. A legal way to take the assets of whoever the stronger side calls guilty is the tool those two conditions were waiting for. Others have walked this road. The United States built civil forfeiture to break drug empires and ended up with police taking ordinary people’s cars and cash on suspicion alone, a practice its own critics came to call policing for profit. A power sold against kingpins rarely stays aimed at kingpins. And the tidy revival of this bill, the out-of-town crowd, the signed letter, the neat deadline, sits strangely beside how the real protests were met, the ones smeared as foreign-funded.
Nevertheless, it is not wrong for pro-reformists to want this bill passed. Corruption bleeds the country, recovery of stolen money is shamefully slow, and safeguards can be written into any text. The safeguard that matters most doesn’t live in text, but in weak institutions that need strength and integrity to rightfully wield the law. A law like this is only as safe as the people who control it.
Rayhan has a background in government affairs and public policy, with experience across government institutions and advisory firms. His work focuses on the intersection of geopolitics, policy, and risk, with expertise in advocacy, regulatory analysis, and stakeholder engagement. He holds a degree in Government from Universitas Padjadjaran, and has completed an exchange at Universitat Pompeu Fabra in Spain, focusing on global politics and sustainability.
Malaysia 🇲🇾
Hosting the AI Race
by Muhammad Aiman Bin Roszaimi, in Cyberjaya
Malaysia is becoming one of Southeast Asia’s most important destinations for artificial intelligence infrastructure. But as billions of ringgit flow into data centres, is Malaysia building its own technological power, or simply becoming the infrastructure that powers someone else’s?
Australia-based Firmus signed a multi-year agreement with OpenAI to provide computing capacity from two Malaysian data centres. The deal makes OpenAI an anchor customer and will see Nvidia’s next-generation Vera Rubin processors deployed across Firmus’ Asia-Pacific operations. Reuters describes Malaysia as Southeast Asia’s fastest-growing data-centre market, although the expansion is increasingly drawing scrutiny over electricity and water consumption.
On the surface, this is exactly the kind of investment Malaysia wants. Data centres bring foreign capital, construction, digital infrastructure, skilled employment, and connections to the global AI economy. They strengthen Malaysia’s position between Singapore and other emerging regional hubs while giving the country a larger role in the infrastructure behind the next generation of AI.
But infrastructure is not the same thing as technological ownership.
Malaysia is reportedly considering Huawei’s Ascend 910C chips for a RM2 billion sovereign AI initiative aimed at giving the country greater control over sensitive national data. The project would be operated through Telekom Malaysia, according to people familiar with the matter, although the government has not publicly confirmed that Huawei has been selected.
This is where Malaysia’s AI strategy becomes strategically interesting. First, how can Malaysia capture more value from the global AI boom? Second, how can Malaysia prevent critical national data and computing capabilities from becoming excessively dependent on foreign technology providers? These are not necessarily the same problem.
A country can host the world’s largest AI companies while remaining dependent on foreign chips, cloud platforms, software ecosystems, and intellectual property. The servers may sit in Malaysia, the electricity may come from Malaysia, and the buildings may be constructed by Malaysian companies. Yet the most valuable technological layers can still belong elsewhere.
This is also why the current debate over Huawei should not be reduced to a simple “China versus America” question. For Malaysia, the deeper issue is “strategic dependence”.
Choosing one foreign technology ecosystem over another does not automatically produce technological sovereignty. It may simply change the source of dependency. True technological sovereignty requires Malaysia to develop capabilities to design, adapt, secure, and eventually own critical technologies rather than merely procure them.
Therefore, the danger is not that Malaysia attracts too much AI investment, but that Malaysia succeeds at attracting AI infrastructure without succeeding at moving up the technological value chain.
Malaysia does not need to become self-sufficient in every layer of the AI ecosystem. That would be unrealistic. But it does need to identify which dependencies are acceptable and which capabilities are too strategically important to outsource indefinitely.
The real test of Malaysia’s AI ambition, therefore, is not whether OpenAI, Nvidia, Huawei, or other technology giants choose Malaysia.
It is what Malaysia becomes because they do. That distinction matters beyond Malaysia, too.
Aiman is a PhD candidate in Security and Strategic Analysis at the National University of Malaysia. His research focuses on Malaysia’s space policy, ASEAN regional security, and the strategic implications of emerging technologies. His work explores how Malaysia’s defense policy and strategic culture shape its approach to outer space.
Editorial Deadline 05/09/2026 11:59 PM (UTC +8)
