Japan's Biggest Intelligence Overhaul Since WWII: What the New Spy Agency Means for Asia's Security Architecture

Japan's parliament moved legislation on 27 May 2026 that would establish a centralised national intelligence agency, a reform that analysts describe as the most sweeping restructuring of the country's secret-state apparatus since the US occupation drafted the postwar constitution in 1947. The bill, backed by the ruling coalition and passing through committee ahead of a full chamber vote, would fold currently dispersed intelligence functions — currently split across the Cabinet Intelligence and Security Intelligence Council, the Public Security Intelligence Agency, and military signals units — into a single civilian-led body modelled loosely on the American CIA structure.
The timing is not incidental. North Korea's accelerating missile programme, China's expanding maritime presence in the East China Sea and Pacific, and Russia's deepening strategic partnership with Pyongyang have collectively sharpened Tokyo's sense of exposure. successive annual defence white papers have documented what Japanese strategists call a "grey zone" problem: challenges to Japanese sovereignty that fall below the threshold of conventional warfare but are persistent, deniable, and escalating. A consolidated intelligence service, its proponents argue, can join those dots faster than a fragmented system that routes signals intercepts through military channels, human intelligence reports through the Foreign Ministry, and domestic threat assessments through the National Police Agency — with no single institution responsible for synthesising the picture.
What the Reform Actually Does
The proposed agency — provisionally titled the External Intelligence Agency in draft legislation — would assume primary responsibility for foreign intelligence collection and analysis, taking that function away from the Japan Self-Defence Forces' intelligence branch and the Cabinet Secretariat's intelligence division. The Self-Defence Forces would retain their tactical intelligence role for military operations. The new body would sit under cabinet authority, with a civilian director appointed by the prime minister and subject to confirmation by both houses of the Diet.
Several provisions address civil liberties concerns that emerged during the legislative consultation process. Intelligence products produced by the agency could not be used directly as evidence in domestic criminal proceedings — a firewall intended to prevent mission creep into law enforcement. An oversight board comprising retired judges and legal scholars would receive classified briefings quarterly. The government has also proposed annual statistical disclosures on the agency's headcount and budget allocation, though critics note these figures will offer limited window into actual operations.
The existing Public Security Intelligence Agency, which monitors domestic extremist groups under the Subversive Activities Prevention Act, would not be absorbed — a deliberate carve-out that keeps domestic surveillance within a separate legal framework with its own court-supervised warrant regime. This distinction between foreign and domestic intelligence collection mirrors the structure of the US Intelligence Community, where the CIA and the NSA operate under different legal authorities than the FBI's intelligence division.
The Accountability Gap
Japan is not alone in Asia in confronting the tension between security effectiveness and democratic oversight. South Korea, which has fought its own battles over the legacy of its authoritarian-era intelligence services, restructured its foreign and domestic intelligence agencies repeatedly across the 1990s and 2000s before eventually consolidating them under the NIS in 2024. The Korean experience offers a cautionary note: concentrated intelligence budgets and reduced inter-agency rivalry can improve policy coherence, but they also concentrate the potential for abuse.
In Japan's case, the risk is compounded by the country's post-war culture of intelligence restraint. The Self-Defence Forces historically operated under significant legal constraints on offensive intelligence activities, and Japan's signals-intelligence capabilities — while advanced in technical terms — have been orientated toward alliance coordination with US services rather than autonomous global reach. A centralised civilian agency with a broader mandate and a direct line to the prime minister's office changes that dynamic. Whether the proposed oversight board has the expertise, the access, and the institutional independence to serve as a genuine check is a question the legislation, as drafted, does not fully answer.
The government's framing has emphasised coordination rather than expansion — the reform as efficiency play rather than power grab. Officials point to the existing intelligence-sharing arrangements with Five Eyes partners, particularly the United States, and argue that Japan's fragmented architecture has at times slowed the flow of finished intelligence to alliance counterparts. A unified service, they contend, would make Japan a more reliable partner in multilateral security frameworks.
That argument is not without merit, and counterparts in Canberra and Washington have signalled quiet support. But it also contains an implicit acknowledgment that Japan's intelligence ambitions have outgrown its Cold War-era institutional design — and that the question now is not whether Tokyo will project more intelligence power, but how that power will be governed.
Regional Reverberations
The reform lands unevenly across the region. Beijing has not issued a formal response, but Chinese state media commentary framed the legislation as further evidence of Japan's drift toward a "normal country" security posture — language used to describe Tokyo's incremental departure from the post-war pacifist settlement. The framing is self-serving, but it is not entirely wrong: each successive defence and intelligence reform since Shinzo Abe's first administration has chipping away at constraints that were, in significant part, designed to prevent exactly this kind of institutional capacity.
For Taiwan, which faces the most acute threat calculus from China's military modernisation, a more capable Japanese intelligence service is likely welcome. Tokyo has steadily deepened its unofficial security cooperation with Taipei — sharing maritime domain awareness, hosting defence attachés, and participating in multilateral exercises that implicitly address cross-strait contingencies. A centralised foreign intelligence agency could formalise channels that currently operate through informal arrangements, potentially improving the quality of strategic signal that Taiwan receives from a regional partner.
South Korea's reaction is more complicated. The two countries have made measurable progress on intelligence sharing since the trilateral summit frameworks established in 2024, but historical grievances rooted in Japan's colonial occupation remain a friction point in any security cooperation. A Japanese intelligence agency with greater reach is, from Seoul's perspective, both a potential asset and a reminder of the asymmetry that characterises the regional order.
What Happens Next
The bill is expected to clear the lower house before the current parliamentary session ends in June, with the upper house vote following. Agency establishment would take an estimated eighteen to twenty-four months — meaning operational capability, at the earliest, in late 2027 or 2028. The appointment of the inaugural director will be the first real signal of the agency's orientation: whether it tilts toward the bureaucratic consolidation model preferred by the Defence Ministry, or toward a more autonomous civilian service with its own analytical culture and collection priorities.
Japan's allies will be watching that appointment closely. So will its neighbours. The reform reflects a country that has decided, after years of deliberation, that it needs to see further and act faster in the intelligence domain. Whether it has built in the right safeguards for a democracy still working out how to balance secrecy and accountability is a question that will take years to answer — and one that the sources currently before this publication do not fully resolve.