AISEAN

ADMM Cybersecurity & Information Centre of Excellence (ACICE); Malware Information Sharing Platform; Annual Digital Defence Symposium (300 officials, 31 nations, 2025); REAIM co-host; SAF AI integration pipeline via MINDEF commercial tooling

Chaired UN OEWG on ICT Security 2021–2025; strongly advocates human-in-the-loop doctrine; ST Engineering Terrex s5 contract highlighted at Singapore Airshow 2026

BrahMos battery deliveries from India (2nd battery 2026); Expanded US-PHL maritime drills focused on domain awareness; AUS-CAN-PH-US multilateral maritime cooperation activities (Aug 2025)

All national agencies ordered to cut non-essential spending amid Hormuz-crisis economic fallout, creating tension with modernization ambitions

Scorpène-class submarine contract (2024); 2× Airbus A400M; Optimum Essential Force modernisation; Presidential AI Regulation expected early 2026

~$5.9B in additional energy subsidies required in 2026 due to Hormuz disruption, squeezing defence headroom

FA-50M light combat aircraft deliveries; Naval Strike Missile (NSM) for KD Jebat frigate; 2× Multi-Role Support Ships; National AI Office with ethics code & 5-year plan; announced sovereign AI stack using DeepSeek (May 2025)

Ministry of Defence contributing to budget-saving measures; Malaysian DeepSeek AI stack deployment raises counterintelligence concerns given US chip-access warnings

Vietnam–Russia nuclear power cooperation (early 2026); updated Cybersecurity Law with AI-incident provisions; National AI Development Fund; growing data centre investment from Google, Microsoft, Amazon

AI Law leaves high-risk activity classification to sectoral bodies; defence ministry's AI posture remains opaque; dual-use risk from civil-military AI infrastructure convergence

Defense & Security 2025 Bangkok (Nov 2025, 580+ exhibitors, LAWS governance seminars); AI workloads at 28% of Thai data centre capacity; ETDA AI law draft under revision; Chinese tech (Huawei, Alibaba, Tencent) fully embedded in data centre market

Recent Cambodia–Thailand border clashes (2025) highlighted near-absence of autonomous weapon deconfliction mechanisms; debt burden may constrain future AI defence procurement

AI-powered surveillance systems—drones with computer vision, sensor fusion, satellite feeds—provide real-time tracking of grey-zone vessels, coast guard incursions, and maritime militia activity across the South China Sea and Strait of Malacca.

Singapore's ACICE runs a live Malware Information Sharing Platform; Philippines integrates AI-assisted cueing in Balikatan exercises; proposed ASEAN Maritime AI Testbed would blend civilian and military feeds starting with Malacca Strait and Sulu Sea

Indonesia's indigenously developed Elang Hitam MALE drone offers 24-hour endurance for persistent surveillance and eventual AI-guided routing and sensor prioritisation; other ASEAN states rely on imported UAV platforms pending domestic development.

US V-BAT drones with Shield AI Hivemind autonomy software (GPS-denied operation) deployed from USS Portland in South China Sea, Jun 2025; Japan's OSA programme now supplies surveillance UAVs and radars to Philippines, Malaysia, Indonesia

China is fielding wave-powered AI-controlled USVs capable of 12+ months continuous operation and 10,000 km range for persistent grey-zone presence around reefs and disputed features. China's 'smart maritime defence system' links air, sea, space, and shore assets into a fused kill-web architecture.

Chinese wave-powered USVs described as 'near silent' and suited for submarine detection, tripwire sensing, and swarm grey-zone operations (May 2026); China's 'Underwater Great Wall' deploys seabed sensors for real-time submarine tracking across SCS; PLA Zhu Hai Yun autonomous mothership conducting trials

Sovereign AI cloud infrastructure enables classified model training, red teaming, and AI decision-support pipelines wired into defence operations. Singapore's Oracle isolated-cloud deal is the region's first such arrangement; other ASEAN states are integrating commercial AI stacks into national command systems.

Singapore–Oracle isolated defence cloud contract (Mar 2025); ACICE Digital Defence Symposium 2025 gathered 300 officials from 31 countries including US and China; Indonesia's Presidential AI Regulation expected to formalise civil-military AI coordination in 2026

Coordinated AI-scaled influence operations targeting ASEAN publics (especially Filipino narratives on maritime claims and US alliances) are active. ASEAN states are building counter-disinformation capacity via ACICE but remain asymmetrically outmatched by state-level adversaries.

Reuters investigation (Oct 2025) documented coordinated influence activity targeting Filipino narratives including local cutouts and cash awards; Chinese AI models at ~30% global usage share by end-2025 (up from 13%), deeply embedded in ASEAN civil/military AI stacks; DeepSeek flagged by US officials as supporting Chinese military/intelligence work while seeking chips via SEA shells

LAWS—including Lancet-style loitering munitions—are becoming cheaper and more accessible. ASEAN militaries are not primary developers but face rapid exposure as these systems diffuse into the region's threat environment, compressing human decision windows to seconds.

Lowy Institute (Sep 2025) warns of LAWS deployment in Cambodia–Thailand-style clashes; ASEAN Defence Ministers' 2025 statement expresses concern about autonomous weapon systems impact on 'regional and international stability'; proposed 3–5 year moratorium on highest-risk LAWS deployments under discussion

Indonesia's Elang Hitam MALE drone achieved first flight mid-2025, offering 24-hour endurance for persistent ISR and eventual autonomous routing. Singapore has embedded drone literacy into basic military training and stood up a dedicated Counter-UAS group. Thailand activated a UAS battalion post-Cambodia clashes.

Elang Hitam: 24-hr endurance MALE; SAF Counter-UAS group operational 2025; Thailand UAS battalion stood up 2025/26; fiber-optic drones used in Cambodia-Thailand border clashes

AI-enhanced MDA is the priority non-kinetic application across ASEAN navies, enabling real-time vessel behaviour analysis, grey-zone tactic detection, and early warning in the South China Sea. China is simultaneously deploying AI-controlled wave-powered USVs for persistent low-cost SCS patrols.

ADMM-Plus EWG tasked with AI-MDA integration; China CSSC research on wave-powered AI USVs published 2026; ASEAN AI MDA flagged in RSIS, Lowy Institute analyses as top priority use-case

Singapore contracted Oracle for an isolated cloud environment for its defence technology arm in March 2025 — the first such arrangement in Southeast Asia — enabling classified AI model training, red teaming, and secure deployment at scale. Other ASEAN states remain on commercial or unsecured infrastructure.

Singapore-Oracle isolated defence cloud: March 2025; described as 'first-of-its-kind in the region'; enables classified model training and red teaming at scale

ASEAN militaries are beginning to integrate AI decision-support systems for ISR fusion, drone identification, and threat prioritisation. The ADMM Joint Statement explicitly commits member states to ensuring 'accountability and responsibility can never be transferred to machines,' setting a human-in-the-loop baseline.

ADMM Feb 2025 Statement: accountability cannot transfer to machines; AI DSS for drone ID/prioritisation noted in The Diplomat May 2026; ADMM-Plus TTX/FTX exercises to incorporate AI elements per 2025 RSIS guidance

The Asia-Pacific counter-drone market is forecast to grow from $1.05B (2025) to $3.89B by 2030 at 30% CAGR. GNSS jamming, AI-based drone classification, interceptor drones, and directed-energy effectors are all entering SEA defence inventories. Israeli (Drone Guard, Drone Dome) and US (Anduril Lattice) systems are key competitors in open SEA markets.

Asia-Pacific C-UAS market: $1.05B→$3.89B by 2030 (30% CAGR); Thai UAS Warfare Center 2026; SAF Counter-UAS group 2025; Israeli & US C-UAS primes active in SEA market (DSA Kuala Lumpur, Milipol TechX April 2026)

AI-scaled influence operations targeting ASEAN publics are an active and documented threat. Reuters (Oct 2025) detailed coordinated influence activity targeting Filipino maritime narratives. The ACICE Malware Information Sharing Platform provides ASEAN states with shared cyber/disinfo threat intelligence as a first-line defence.

Reuters Oct 2025: coordinated influence ops targeting Philippines maritime/vaccine narratives; ACICE platform provides ASEAN-exclusive tactical threat data; DeepSeek flagged by US officials as supporting PLA intel while seeking chips via SEA shells

Multiple SEA states now operating or fielding medium-altitude long-endurance drones for persistent South China Sea ISR, replacing expensive crewed patrol sorties.

Malaysia: 3 × ANKA-S (30+ hr, 30,000 ft, satellite-controlled) at Labuan; Indonesia: Elang Hitam first flight mid-2025 (24-hr endurance); Philippines: US MQ-9A Reapers deployed; China: Wing Loong-2 (32-hr) on continuous SCS patrols

AI-augmented command-and-control tools that fuse ISR feeds and provide real-time decision recommendations to commanders, with human-in-the-loop mandated by REAIM norms.

Singapore unveiled ADSS at Exercise Forging Sabre 2024; Philippines and Indonesia signalled interest in AWS/ADSS acquisition; ADMM joint statement emphasises human oversight as prerequisite

China is deploying AI-driven unmanned surface vessels (USVs) and underwater drones (UUVs) for persistent SCS presence; SEA states are monitoring but not yet fielding comparable systems.

China: 5+ UUV designs including AJX002 (pump-jet stealth) and UUV-300 (torpedo/mine capable); wave-powered AI USVs proposed for law-enforcement patrols; Zhu Hai Yun crewless drone carrier with AI Ocean Stereo Observing System

ASEAN militaries are integrating AI for threat detection, malware sharing, and counter-disinformation, underpinned by Singapore's ACICE and the ADMM Malware Information Sharing Platform.

ACICE Malware Platform provides all 10 ASEAN states tactical/technical cyber threat feeds; coordinated influence operations targeting Filipino SCS narratives documented (Reuters, Oct 2025); US Cyber Command 2026 budget funds AI roadmap for cyber superiority

US-led expansion of MQ-9 ISR architecture across Indo-Pacific knits allied SEA partners into a shared intelligence web, raising both deterrence value and escalation sensitivity.

MQ-9 network now described as 'collective ISR web' combining long-endurance reconnaissance with strike options; Philippines–US expanded maritime drills 2025 focused on contested-water interoperability; China's SCSPI think tank flagged MQ-9 as primary US reconnaissance mainstay against PLA

Ukraine war lessons are accelerating SEA interest in loitering munitions with autonomous targeting features; no SEA state has publicly fielded LAWS but acquisition signals are multiplying.

Philippines and Indonesia formally signalled AWS acquisition interest; ASEAN's 58th AMM communiqué expressed concern over autonomous weapon systems; ADMM urged human control standards before deployment; UN Secretary-General called for binding LAWS treaty with 2026 target

Loitering munitions and ground robots are becoming cheaper and easier to field across Southeast Asian flashpoints. The window for human intervention in autonomous encounters can narrow to seconds.

PLA tested AI drone swarm of ~200 UAVs supervised by one soldier (Jan 2026). China's Wing Loong-2 operates up to 32 hours over SCS. ASEAN has no binding LAWS restraint instrument as of Jun 2026.

China is developing AI-controlled wave-powered USVs for persistent, near-silent maritime law enforcement and surveillance, capable of acting as grey-zone militia proxies in disputed waters.

China's Zhu Hai Yun AI drone carrier (88.5m, deploys 50+ aerial/surface/underwater drones, ~100mi range) operated in a Philippine sea zone without authorization (Jul 2025). US funding allocated for a Philippine Navy USV base.

Long-endurance UAVs are becoming the primary persistent ISR tool across the SCS, with ASEAN states accelerating both acquisition and domestic production, pulling in autonomy for routing and sensor tasking.

Indonesia's Elang Hitam MALE drone (24h endurance) first flight mid-2025. China's WZ-9 Divine Eagle deployed from Hainan. Vietnam and Indonesia both developing/acquiring drones per SCMP Dec 2025.

ASEAN militaries are moving beyond generic AI to sovereign-adjacent, classified-capable platforms enabling secure model training, red teaming, and integration with operational data pipelines.

Singapore contracted Oracle for isolated defence cloud & AI (Mar 2025) — first such deal in the region. Malaysia launched ILMU LLM (Aug 2025). Singapore's SEA-LION pre-trains on 13 regional languages.

Chinese military drones have demonstrated transponder spoofing over the SCS, transmitting false aircraft identities — a likely test of electronic deception relevant to grey-zone and kinetic scenarios.

Reuters (Feb 2026): Chinese military drone (callsign YILO4200) made 23+ flights over SCS since Aug 2025 transmitting false transponder IDs. PLA AI enhances OODA loop speed in cyber and space per ASPI (Apr 2026).

Issued February 2025, the statement encourages ADMM-Plus expert groups to incorporate AI and coordinate efforts. The 58th AMM Joint Communiqué (under Malaysia's ASEAN chairmanship) reinforced it, expressing concern about autonomous weapon systems' impact on global and regional stability. Critics note it is voluntary, leaves operational definitions open, and has no verification mechanism.

Published in 2025 alongside the earlier ASEAN Guide on AI Governance and Ethics, the Roadmap sets foundations for trust and responsible AI adoption by ASEAN member states. Singapore's MINDEF described these as 'critical foundations' at the Seoul Defence Dialogue in September 2025.

Vietnam's AI Law introduces risk-based AI system classification, a National AI Development Fund, and regulatory sandbox mechanisms. It was passed alongside updated cybersecurity and IP laws with AI-specific incident provisions. Sectoral bodies determine high-risk classification, including defence-adjacent applications.

Launched by Singapore in 2023, ACICE promotes multilateral information sharing and capacity building on cyberattacks, disinformation, and AI threats. It operates the Malware Information Sharing Platform (exclusive ASEAN access) and hosts annual Digital Defence Symposiums. In 2025 it convened 300 defence officials from 31 nations including US and China.

Proposed in September 2025, the framework would mandate human veto for lethal action, require test notifications and an incidents hotline, and impose a 3–5 year moratorium on the riskiest LAWS deployments. It would allow entry into force with a threshold of ratifiers, letting committed states set regional norms ahead of global treaties.

Adopted at ADMM Retreat, Penang, 26 February 2025. Commits all 10 ASEAN defence ministers to responsible and accountable AI use, explicitly stating that 'accountability and responsibility can never be transferred to machines.' Directs ADMM-Plus EWGs to incorporate AI. Remains voluntary with no verification mechanism.

Published by ASEAN (2024/2025), this is the region's first coordinated AI governance effort. Analysts at the Lowy Institute note the voluntary guidelines notably omit military AI regulations — a gap that critics argue must be filled by a separate, defence-specific code of conduct.

Published in 2025 under Malaysia's ASEAN chairmanship. Noted by Singapore's Senior Minister of State for Defence as a 'critical foundation for trust and responsible adoption of AI.' The 58th AMM Joint Communiqué (2025) also expressed concern over 'negative consequences of autonomous weapon systems on global and regional stability.'

Launched by Singapore in 2023, ACICE promotes multilateral information sharing on cyberattacks and disinformation. Its Malware Information Sharing Platform gives ASEAN states exclusive access to tactical cybersecurity intelligence. In 2025 it convened 300 defence officials from 31 countries — including the US and China — at the Digital Defence Symposium.

A September 2025 Lowy Institute analysis calls for ASEAN to pursue a narrow, time-bound agreement banning fully autonomous lethal targeting without human oversight, mandating a human veto on lethal action, and establishing a 3–5 year moratorium on the riskiest LAWS deployments. Modelled on the Southeast Asia Nuclear Weapon-Free Zone (SEANWFZ) framework.

The ADMM February 2025 Joint Statement explicitly references UNGA Resolution 79/239 on 'AI in the military domain and its implications for international peace and security,' anchoring ASEAN's regional commitments to the broader multilateral legal architecture and the UN Secretary-General's call for a legally binding LAWS treaty by 2026.

Issued February 2025, the statement encourages ADMM-Plus expert groups to incorporate AI and coordinate AI leveraging efforts. It is voluntary and leaves operational definitions and verification open, but Malaysia's ASEAN chairmanship elevated it into the 58th AMM foreign ministers' communiqué.

Enacted December 2025, effective March 2026. Covers 36 articles with three-tier risk classification, transparency requirements, AI incident management, and mandatory human oversight for high-risk systems. Foreign AI providers must appoint a legal representative in Vietnam.

The Working Group on AI Governance (WG-AI) published an expanded edition in January 2025 covering Generative AI. It is non-binding but forms the baseline for national frameworks. The upcoming DEFA agreement may formalise some principles as binding cross-border standards by end-2026.

Adopted at the 6th ASEAN Digital Ministers' Meeting (ADGMIN) in Hanoi, January 2026, under theme 'From Connectivity to Connected Intelligence.' ADM 2030 sets ASEAN's strategic direction on digital transformation and frames engagement with dialogue partners including the US.

Established in 2025, the ASEAN AI Safety Network serves as a regional platform for capacity building, regulatory preparedness, and safeguard measures. Identified as a key initiative to watch in 2026 under the Philippines' ASEAN chairmanship digital priorities.

The UN Secretary-General's New Agenda for Peace called for a legally binding treaty prohibiting LAWS from operating without human oversight, with a 2026 target. Member states reaffirmed dialogue at the Summit of the Future but efforts remain splintered and largely non-binding.

Enacted December 2025, effective March 2026. Covers 36 articles including risk-based classification, AI incident management, and state oversight. High-risk AI systems require impact assessments and human oversight mechanisms. Foreign AI providers must appoint a local legal representative.

Adopted at the 6th ASEAN Digital Ministers' Meeting (ADGMIN), Hanoi, 15–16 Jan 2026. Ministers endorsed the ASEAN AI Safety Network (established 2025) for capacity building and regulatory preparedness. Secretariat based in KL. Philippines' 2026 ASEAN chairship aligns priorities around AI cooperation and cyber resilience.

Published by the ASEAN Working Group on AI Governance (WG-AI) in 2025. Provides risk-based, voluntary ethical principles for AI deployment. All 10 AMS now have some AI strategy in place (Brunei released its guide in late 2025). The framework is voluntary and leaves defence/dual-use applications largely to member states.

Advocated by Lowy Institute (Sep 2025) in response to Cambodia–Thailand border incidents and SCS drone proliferation. Proposal: mandate human veto for any lethal action; 3–5 year moratorium on riskiest LAWS deployments; basic transparency measures including test notifications and incidents hotline. 58th AMM communiqué (2025) expressed concern about autonomous weapons' impact on regional stability.

Singapore chaired the UN OEWG on Security of and in the Use of ICTs (2021–2025), producing substantive discussion on emerging threats and international law in cyberspace. Singapore also actively supports the Responsible AI in the Military Domain (REAIM) process as a regional champion for responsible military AI.

China is developing wave-powered, near-silent AI-controlled USVs capable of year-long autonomous patrols around disputed reefs and islands at a fraction of conventional naval cost. Combined with the 'Underwater Great Wall' seabed sensor network and AI submarine-detection claims (Sep 2025 — 5% sub survival rate cited), China is systematically compressing the decision space available to ASEAN states and US partners. ASEAN has no doctrine or agreement to govern encounters with such systems.

The Cambodia–Thailand border clashes of 2025 demonstrated how quickly intra-ASEAN conflicts can escalate. Introducing loitering munitions and ground robots into similar future engagements could compress detection-to-detonation timelines to seconds with no space for political intervention. The February 2025 ADMM statement is voluntary and contains no operational definitions, verification, or incident hotline — leaving the region exposed.

Chinese AI models held ~30% of global usage share by end-2025, with deep penetration into ASEAN civil and emerging military AI infrastructure. Malaysia announced its sovereign AI stack uses DeepSeek (May 2025). US officials have simultaneously flagged DeepSeek as supporting Chinese military and intelligence work and seeking chips through Southeast Asian shell companies — creating acute counterintelligence and supply-chain compromise risks for ASEAN defence planners.

The strength of US alliances and security treaties that preserved regional stability for decades is assessed as fading precisely as territorial disputes sharpen. US INDOPACOM still describes AI as a 'force multiplier' but Trump-era deregulation and tariff shocks are undermining trust. The resulting strategic vacuum is accelerating uncoordinated national arms buildups, increasing the risk of miscalculation rather than deterrence.

A Reuters investigation (Oct 2025) documented coordinated influence operations targeting Filipino narratives on maritime claims, vaccines, and the US alliance using local cutouts and cash incentives. AI-scaled content generation will amplify such operations in the next cycle. ASEAN states lack coordinated AI-detection and counter-narrative infrastructure at comparable scale, creating exploitable cognitive vulnerabilities in frontline states most exposed to Chinese coercion.

China's South China Sea island outposts now provide 'all-weather, all-round situational awareness' per a July 2025 PLA article, integrating SIGINT, EW, and anti-ship/anti-air missile systems. The AI-enhanced Qinzhou-class frigate conducted combat drills in the SCS in 2025. China may have surpassed the US in AI-enabled drone swarm capability according to a Tamkang University analyst, and is developing wave-powered AI USVs for persistent SCS patrol — directly threatening ASEAN EEZ sovereignty.

Sustained Chinese Coast Guard and maritime militia operations at Second Thomas Shoal and Scarborough continue in 2026. A Reuters investigation (Oct 2025) documented coordinated AI-scaled influence operations targeting Filipino public opinion on maritime claims, vaccine narratives, and the US alliance using local cutouts. The US and Philippines expanded Balikatan 2026 with a heavy focus on contested-waters interoperability, raising Chinese counter-pressure.

Bloomberg ship-tracking data (May 2026) shows a sharp increase in Chinese vessel incursions within 10 nautical miles of Vietnamese-claimed waters compared to 2022–23. Vietnam's deterrence build-up — US patrol boats, coast guard cutters, diversified security partnerships — signals intent to contest Chinese encroachment, while analysts warn 'the odds of a confrontation are rising even as both countries seek to warm ties.'

Cambodia's use of fiber-optic drones against Thai forces in 2025 border clashes was Southeast Asia's first documented intra-regional UAS combat engagement. Thailand responded by establishing a UAS Warfare Center and a dedicated UAS battalion. The episode exposed the absence of any ASEAN-level norms or hotlines for drone incidents, and demonstrated how rapidly drone escalation can occur between member states.

US officials have named Chinese firm DeepSeek as supporting PLA military and intelligence work while seeking advanced chips through Southeast Asian shell companies. This raises acute compliance and counterintelligence risks for ASEAN economies — particularly Singapore, Malaysia, and Vietnam — that are actively courting cloud, semiconductor, and AI infrastructure investment, creating potential vectors for illicit technology transfer to Chinese defence programmes.

The February 2025 ADMM Joint Statement is voluntary and leaves operational definitions of human control and autonomous engagement undefined and unverifiable. Analysts at RSIS and the Lowy Institute warn this is 'a floor, not a ceiling.' Meanwhile, LAWS procurement is accelerating across the region through imported hardware pre-loaded with foreign doctrine — risking ASEAN states inheriting engagement thresholds and autonomy levels set by great-power suppliers, not by Southeast Asian strategic culture.

China's Wing Loong-2 UAVs conduct 32-hr continuous SCS patrols while deploying AI-guided UUVs (AJX002, UUV-300) and wave-powered USVs. Malaysia's ANKA-S, Indonesia's Elang Hitam, and US MQ-9s operated from the Philippines create overlapping autonomous ISR corridors with no agreed rules of engagement or hotline for AI incidents — raising misperception and accidental escalation risk.

The ADMM's February 2025 joint statement is voluntary and leaves 'human control' undefined. As SEA states acquire semi-autonomous systems, there are no enforceable minimum standards for weapons release authority, model-specific event logs, or cross-border AI incident hotlines — exactly the gap that Lowy Institute analysts warn could enable autonomous escalation with narrowing windows for political intervention.

The US expansion of MQ-9 Reapers across the Indo-Pacific creates a 'collective ISR web' that the PLA's SCSPI think tank characterises as a primary reconnaissance threat, reportedly countered with air defenses, jamming, and comparable Chinese unmanned systems. SEA states hosting US drone operations face dual pressure: enhanced domain awareness versus Chinese political and economic coercion.

Reuters documented coordinated AI-amplified influence operations targeting Filipino public narratives around South China Sea disputes in October 2025. With AI-generated disinformation scalable and cheap, all SEA claimant states face manipulation of domestic political support for maritime postures — a direct national security vulnerability feeding into crisis decision-making.

Only Vietnam has enacted binding AI legislation (effective March 2026). Indonesia is finalising mandatory requirements, Malaysia and Thailand have draft frameworks, while Cambodia, Laos, and Myanmar remain ungoverned. Divergent national rules create compliance arbitrage risks for dual-use AI systems and undermine the ASEAN-level soft-law architecture before it consolidates.

Thailand's restarted S26T submarine contract with China's CSOC (replacing German engines), China's dominance in commercial drone components, and the presence of Chinese-origin AI systems in SEA civil infrastructure create latent dependency vectors. As SEA militaries integrate AI into command systems, supply-chain provenance and backdoor risks in Chinese-origin hardware and software are insufficiently audited.

Reuters reported (Feb 2026) that a Chinese military long-range UAV (YILO4200) conducted 23+ flights over the SCS since Aug 2025, transmitting false transponder identities — including mimicking a sanctioned Belarusian cargo plane and a British Typhoon fighter jet. Operations were concentrated near the Paracel Islands and along Vietnam's coast. Analysts assess this as grey-zone electronic deception testing, directly relevant to future conflict escalation management and raising alarm in Hanoi and Manila.

China's 88.5-metre autonomous drone carrier — framed officially as a civilian ocean science vessel but capable of deploying 50+ aerial, surface, and underwater drones across a ~100-mile range — operated in a Philippine exclusive economic zone without authorisation in July 2025. Analysts warn that data collected could support PLA submarine navigation and anti-submarine warfare. Manila protested; incident intensified Philippine urgency to host US unmanned surface vessel infrastructure.

US officials identified Chinese AI firm DeepSeek as supporting PLA military and intelligence work while allegedly seeking advanced chips through Southeast Asian shell companies. This directly implicates ASEAN markets — particularly Malaysia, which in May 2025 announced its sovereign AI stack uses DeepSeek — in US export-control compliance and counterintelligence risk. The episode raises the stakes for all ASEAN states courting Chinese cloud and semiconductor investment alongside US alliance obligations.

The 2025 Cambodia–Thailand border skirmishes — resolved by direct leadership contact — illustrate how loitering munitions and ground robots could shrink human decision windows to seconds in future incidents. In the South China Sea, coast guard units, naval forces, maritime militia, and civilian traffic create an environment where AI-assisted autonomous systems risk triggering unintended escalation before political leaders can intervene. ASEAN has no binding instrument to govern such deployments as of June 2026.

A Reuters investigation (Oct 2025) detailed coordinated influence activity targeting Filipino public opinion on vaccines, South China Sea maritime claims, and the US alliance — using local cutouts and cash payments. Analysts assess that AI-scaled content generation and synthetic media will dramatically amplify such operations in the next cycle, eroding public support for Philippine security postures and US alliance legitimacy at a moment of peak SCS tension.

The US/Israel–Iran conflict disrupted Strait of Hormuz flows, forcing ASEAN energy-import-dependent economies into difficult trade-offs. Indonesia estimated USD 5.9B in additional energy subsidies; Thailand borrowed USD 12.2B; Malaysia's defence ministry was asked to contribute to savings; the Philippines told agencies to cut non-essential spending. The World Bank projects ASEAN growth slowing from 5.0% (2025) to 4.2% (2026), threatening the multi-year modernisation budgets that underpin AI-defence programmes across the region.