Hello, this is Hamamoto from TIMEWELL.
In mid-July 2026, a New York Times investigation—“How Putin Turned Japan Into a Den of Spies”—spread rapidly in Japan. Yahoo! News expert coverage amplified the shock: Russian military intelligence allegedly using Tokyo as a base, Japanese-made parts showing up in weapons used against Ukrainian cities, and diplomatic warnings that appeared to move too slowly.^[1]^[2]^[3]
Anger is understandable. From an export-control practitioner’s seat, the colder story is structural. This is not only about “bad companies shipping straight to Russia.” It is about lawful-looking trade that still ends on a battlefield—via Vietnam and other third-country hubs, weak red-flag discipline, and a habit of treating Japan’s Foreign Exchange and Foreign Trade Act (FEFTA) list checks as the whole job.
I believe this episode concentrates Japan’s export-control design problem. Below I unpack what was reported, why notice came late, how counter-espionage debates intersect with trade compliance, and what companies can change on Monday morning—not as panic, but as operating design.
If you want a quick self-check first, start with our free export-compliance diagnostic. Even the short version surfaces where your process is thin.
What was reported: a procurement hub, not only direct sales
The Times investigation, based on interviews with intelligence and government officials across continents, describes GRU activity using Japan as a technology-procurement base. Coverage highlights an Aeroflot office in Tokyo allegedly serving as cover, walking distance from Japan’s national police headquarters—the “under their noses” framing that made the story viral.^[1]^[2]^[3]
One figure travels especially fast: Ukrainian government estimates that about 90% of Russian missiles and drones contain Japanese components.^[1]^[3] Read that carefully. It does not mean every part is Japanese. It means wreckage analysis keeps finding Japanese electronics and industrial devices. That is still damning for a country that joined Western export restrictions.
A second axis is diplomacy. Reporting says Ukraine sent multiple diplomatic notes to Japan’s Foreign Ministry from April 2025 onward—about sixteen over a year—documenting concern about Japanese parts feeding Russian weapons.^[2]^[4] That is a state-level failure mode if true. For companies, the quieter point is that the raw materials of risk were already public: METI’s Russia sanctions pages, third-country diversion guidance, and G7 industry guidance on evasion.^[5]^[6]
METI has repeatedly expanded bans and licensing for Russia-related trade—regime-controlled items, dual-use goods useful for military capability, industrial-base items, and more.^[5] It also tells exporters to be careful when shipping Russia-banned items to non-Russia destinations, pointing to Common High Priority Items (CHPI) and related materials.^[5]^[6] A September 2025 METI deck walks through diversion risk and insists on end user, end use, and re-export thinking.^[6]
So split the problem.
One track is counterintelligence: spies operating in Japan, cultivating logistics and commercial contacts. The other is export control and supply chains: Japanese firms unknowingly feeding military demand. They connect. But the lever a company can pull this week is mostly the second track. Brand names of major Japanese electronics makers surface in public debate because Japanese industrial components sit deep in global supply chains—not because every mention equals proven intentional military supply by a named firm. Public government materials focus on the structure of diversion, including third-country routes.^[6]
Replace siloed classification work with AI.
METI's FY2024 data shows 52% of foreign exchange law violations stem from classification errors. Download the TRAFEED product catalog covering features and rollout.
Why “we complied with FEFTA lists” is not enough
Japan’s security export control stack, simplified, is list control (spec-based licensing lists), catch-all controls (end-use/end-user concerns), and the post-2022 Russia-related special prohibitions and approvals.^[5] Many teams still start—and stop—with “is it on the list?” If not, and if the ship-to is not Russia, the file feels clean.
That design has holes.
Dual-use reality. Parts recovered from weapons are not always exotic list items. Capacitors, connectors, sensors, industrial controllers—civilian catalog goods—matter. Russia sanctions deliberately reach into that dual-use band.^[5]^[6] “Not list-controlled, therefore free” fails more often under this regime.
The end user you cannot see on the first invoice. The named customer may be a Vietnamese, Turkish, UAE, or Central Asian trading company. Documents look legal. The risk is resale. METI materials keep drawing Japan → third country → Russia (or Russia-linked military procurement).^[6] U.S. and EU entity designations also hit third-country operators, including cases involving Vietnam among other jurisdictions.^[6] That is not a brief against a whole country. It means logistics near military demand get packaged as ordinary third-country corporations.
Red flags on paper, not in operations. BIS has long published “Know Your Customer” and red-flag style guidance: cash deals, vague military-capable orders, reluctance on end-use certificates, murky re-export stories.^[7] Japanese manuals often copy the list. Busy sales cycles still override it. Paper compliance and floor compliance diverge.
Japan-only screening. EAR can still attach when U.S.-origin content or re-export facts appear. G7 industry guidance and CHPI lists tell exporters which items are especially diversion-prone.^[5]^[8] METI echoes that message.^[5] Yet many classification sheets still end at Japanese list checkboxes.
Incentive asymmetry. Booking revenue hits this quarter. Diversion scandal may hit years later. If stopping a deal is career-negative, red flags die.
The core failure mode, in my view, is design more than cartoon villainy. Export control treated as “permit paperwork” rather than “stop military diversion across the chain” leaves Japanese marks on wreckage.
How diversion is built: ordinary third-country companies
METI’s risk materials map patterns that look ordinary from the shipper’s desk.^[6] A Japanese maker sells to a third-country distributor or regional plant. Inventory sits. Another buyer appears. On paper, Japan never shipped to Russia. From the military buyer’s side, the goal was never “buy from Japan directly.” It was “obtain Japanese-quality parts.” That demand structure explains why Japanese industrial brands keep appearing in open-source wreckage talk—even when intent at the first export node is not proven.
Designated entities in U.S./EU/Japan lists have included operators across Turkey, the UAE, India, Central Asia, and Southeast Asia (including Vietnam in some rounds).^[6] Do not bulk-label countries. Do not relax diligence because the destination “is not Russia,” either.
Group companies create another blind spot. Shipping to a related party “because we know them” can be worse under sanctions: relatedness can increase end-use risk. Logistics noise—generic descriptions, missing use statements, multiple consignees—adds up into a corridor after the fact. Export control is a data continuity job from order entry through after-sales, not a single customs form.
One more limit of public debate: viral brand names on social media are not the same as a proven intentional supply case. Parallel trade, stock resale, and long chains exist. This article’s focus is whether Japan’s export-control design matches a world where industrial parts are hunted for war.
Why notice came late—and where anti-spy policy intersects
Information asymmetry. Wreckage teardown is done by Ukrainian and allied specialists. Japanese compliance teams have shipment data, not battlefield photos. Without diplomatic and intelligence sharing, firms learn late—which is why repeated Ukrainian notes matter.^[2]^[4]
The success-story trap. Direct Japan–Russia export values fell hard after 2022 on METI’s trade-stat framing.^[6] That looks like sanctions working. If third-country shipments rise and feed the same demand, military supply continues. Analyses of Russia’s wartime industrial adaptation via diversion are not new.^[9] “Russia revenue went to zero, so we are safe” is a comforting error.
Organizational splits. Sales owns revenue, logistics owns lead time, compliance owns licenses, legal owns policies. Nobody owns “will this order sit in a missile in six months?” Red flags need authority for sales to stop and for compliance to reverse. Weak compliance becomes a document-fixer after the fact.
The gap between counterintelligence and trade control. Japan long lacked a broad anti-espionage criminal statute of the type people casually call a “spy prevention law.” The Act on the Protection and Utilization of Important Economic Security Information (security-clearance framework in the economic-security field), in force since May 2025, strengthens protection of designated government information.^[10]^[11] It is not a general export-transaction ban. Cover identities that cultivate logistics contacts still require company-side visitor control, partner screening, and tech-data exit controls.
Peace-country bias. For years many firms budgeted export control as cost center paperwork. U.S.-style multi-hundred-million enforcement headlines are not Japan’s daily news diet. War feels far from the warehouse. Distance dulls red-flag sensitivity. That is allocation, not personality: one overworked part-timer cannot investigate beneficial ownership on every third-country account. If headcount is thin, management must narrow the product/customer set—not pretend infinite diligence is free.
Until state counterintelligence, corporate export control, and allied information sharing lock together, each company must look beyond its first ship-to line. Late notice is a design outcome.
What not to confuse in the spy debate
Clearance law advances information security for government-shared secrets.^[10] It does not replace FEFTA screening for commercial shipments. Waiting for police is not a sales-ops plan. Export control is part of economic security—not the whole of counterintelligence. We covered the legal landscape for business readers in our spy-prevention overview. For today’s outbound parts, the immediate lever remains diligence and red flags.
What to change: published HS codes (CHPI), red flags, process
The “battlefield parts list” is public—and coded in HS
Japan’s METI published Common High Priority Items (CHPI) on 20 October 2023 (updated 22 February 2024): items identified by partner governments from Russian weapons recovered in Ukraine, each with a 6-digit HS code, organized in Tier 1–4, totaling 50 items, all already subject to Japan’s Russia export bans.^[14]^[15] METI simultaneously warns that even shipments of those banned categories to non-Russia destinations require careful checks of final destination, end use, and end user—and that routing Russia-banned goods to Russia via third countries can violate FEFTA.^[14]
G7 industry guidance (METI hosts a Japanese provisional translation) states the list covers items recovered from weapons or assessed as essential to Russia’s military production, to help industry due diligence against third-country diversion.^[16]
| Tier | Meaning (official framing) | Example HS codes |
|---|---|---|
| 1 | Integrated circuits critical to precision munitions | 8542.31, 8542.32, 8542.33, 8542.39 |
| 2 | Comms / navigation / preferred G7 electronics | 8517.62, 8526.91, 8532.21, 8532.24, 8548.00 |
| 3.A | Diodes, transistors, cameras, connectors, etc. | 8541.10, 8541.21/129, 8525.89, 8536.69… |
| 3.B | Mechanical / optical / aircraft parts | 8482.10–250 bearings, 8807.30, optics/navigation 9013/9014 |
| 4.A | Design, manufacture, test gear for electronics | 8486.10/20, 8534.00, 9030.x… |
| 4.B | CNC machine tools (highlighted in Feb 2024 update) | 8457.10, 8458.11/891, 8459.61, 8466.93 |
METI’s companion “examples requiring caution” track classic red flags: use vs business mismatch, vague end use, freight forwarder as end destination, Russia branches or counterparties, sanctioned counterparties, post-Feb-2022 first inquiries or volume spikes, oversized quantities, opaque routes, above-market pricing / full prepayment, unrealistically short lead times.^[14]^[15] G7 guidance adds HS misclassification, post-arrival HS switches, shell addresses, residential “offices,” websites scrubbed of Russia links, and more.^[16]
So the gap is not “no list existed.” The gap is operations. Map your SKUs to these HS bands; force enhanced review on third-country sales of hits; escalate automatically on new account + vague use + agent channel.
Turn red flags into records, not vibes. Pause and escalate; do not leave “sales thinks it’s fine” as the control.^[7]
Contract for end use, end user, re-export. Clauses will not stop every liar. They reduce “we never asked.” Ask distributors for periodic resale reporting. Train sales to explain why: under current geopolitics, Japanese exporters are expected to own end-use responsibility. Honest customers usually get it. Customers who hate the question deserve deeper review.
Include technology transfers. Drawings, parameters, software, remote maintenance views, shared cloud folders—these can be controlled service transactions under FEFTA, including deemed export issues.^[12] If the reporting also involves tech theft and procurement ops,^[1] know-how exits belong on the same board as parts.
Stop living on exact-string list hits. Sanctions and user lists fail under transliteration and affiliates. Ownership, shared officers, shared addresses, shared email domains are weak signals that only work when bundled—hard by hand at scale.
A fictional ordinary day
A mid-sized Osaka industrial-parts maker gets a small order from a Southeast Asian agent: sensors and control boards, “line upgrade,” L/C payment. List screening says non-listed. Destination is not Russia. No exact hit on the agent name. Approve. Ship. Book revenue. Three months of copy-paste orders follow. Six months later, a wreckage report shows parts that look uncomfortably familiar. Sales blames the agent. Compliance says “non-listed, third country.” Management says media overreach. Nobody lied. Nobody chased end use. Where could it have stopped? First order depth, resale clauses, CHPI pathing, pattern detection on repeat agent orders. Red flags live in empty fields on order forms—not only in spy novels.
Operating checklist (compressed)
- Management risk appetite. Explicitly back “no-sale” on diversion-suspect third-country deals.
- SKU mapping for lists, catch-all, Russia bans, CHPI-like goods—including OEM white-label.
- Customer and route visibility: beneficial ownership, affiliates, historical resale, shipping patterns.^[6]
- Order mini-reviews when small size meets dual-use meets new customer meets vague use.
- Records and training. Keep decision trails; refresh training when sanctions change; put METI/CISTEC Russia pages on the intranet top.^[5]
- Incident hygiene. If diversion is suspected, preserve data, stop suspect channels, engage authorities as needed—do not rush to “not us.”
- SME vs large enterprise. SMEs should narrow scope and deepen review. Large groups should police overseas sales edges with common policy and audit.
Common pushbacks
“Our parts can’t be military.” Dual-use means they can be used, not that they were designed as weapons.
“The distributor is at fault.” Reputational and banking questions still hit the brand.
“Competitors sell it.” That is not a legal or ethical permission slip.
“Stopping everything kills manufacturing.” The job is selection, not autarky.
“The state should fix it.” States must—but the ship button is corporate.
For how TRAFEED approaches classification, screening, and relationship visibility without replacing the compliance officer, see the TRAFEED overview.
Building organizations that notice—TRAFEED and the first step
TRAFEED is TIMEWELL’s AI agent specialized in export control. In joint validation with Okayama University on roughly 30,000 past review cases, we observed 95%+ AI assessment accuracy (company research). Core risk logic is protected under Japan Patent No. 7862062. Final classification decisions remain with the customer’s export-control officer.^[13]
Where it helps diversion defense: cross-jurisdiction list/sanctions awareness as rules move; customer and affiliate risk visibility beyond exact name match; and explainable records for why a deal stopped or proceeded. AI should accelerate evidence for humans—not replace accountability. A classic failure—third-country new account, CHPI-like SKU, vague end use, sales pressure—should be hard to push through when flags force workflow holds. If holds do not exist, software is décor.
Tools will not arrest GRU officers. They can raise the odds that ordinary commercial loopholes close. That is the outcome I want after this news cycle.
If you only need temperature, use the diagnostic. If you want to redesign process, contact us about TRAFEED.
Turn embarrassment into redesign energy
Headlines will cool. Shipment logs will not. Direct Russia sales falling was never proof that military demand lost Japanese inputs.^[5]^[6] Pride in Japanese manufacturing and higher-resolution caution must travel together.
One sentence for the internal memo:
Stopping direct Russia shipments is not enough. Check whether ordinary third-country deals still sit on military demand. Red flags are procedures, not vibes. Embed end use, end user, and re-export into contracts and reviews. Late notice was design failure—and design can change.
Measure the gap first—free diagnostic—then talk if you want system change.
References
[^1]: The New York Times, “How Putin Turned Japan Into a Den of Spies,” 2026-07-12. https://www.nytimes.com/2026/07/12/world/asia/russia-spies-japan-war-drones-electronics.html
[^2]: Makiko Iizuka, Yahoo! News Expert commentary on the Times reporting, July 2026. https://news.yahoo.co.jp/expert/articles/3b0fb5050b54acd86b71e57437ee7eac7be53e0d
[^3]: NHK World, “NYT: Russia uses Japan to procure high-tech goods for its war,” 2026-07-13. https://www3.nhk.or.jp/nhkworld/en/news/20260713_04/
[^4]: Nikkei reporting on Ukrainian diplomatic notes (as cited in expert coverage, 2026-07-13). https://www.nikkei.com/article/DGXZQOGM1333N0T10C26A7000000/
[^5]: METI, Russia-related sanctions portal. https://www.meti.go.jp/policy/external_economy/trade_control/01_seido/04_seisai/crimea.html
[^6]: METI Trade Control Department, “Risks of Russia sanctions and the need for response” (Sept 2025). https://www.meti.go.jp/policy/external_economy/trade_control/01_seido/04_seisai/downloadCrimea/20250912_risk.pdf
[^7]: U.S. BIS, Know Your Customer / red-flag style guidance. https://www.bis.gov/
[^8]: U.S. BIS, Common High Priority Items and related Russia export-control materials. https://www.bis.gov/
[^9]: Nikkei, Russia defense industry sustained by wartime mobilization and diversion (2024-02-23). https://www.nikkei.com/article/DGXZQOGM051A70V00C24A2000000/
[^10]: Cabinet Office, Act on the Protection and Utilization of Important Economic Security Information. https://www.cao.go.jp/keizai_anzen_hosho/hogokatsuyou/hogokatsuyou.html
[^11]: Practice commentary on the clearance act (in force 2025-05-16). e.g. https://www.businesslawyers.jp/articles/1402
[^12]: METI security export control (including deemed export / services). https://www.meti.go.jp/policy/anpo/
[^13]: TIMEWELL TRAFEED product information (Patent 7862062; Okayama University joint validation; ~30k cases; 95%+ accuracy, company research; final decision by customer). https://timewell.jp/trafeed
[^14]: METI, notice on Russia export bans and CHPI (20 Oct 2023; partial revision 22 Feb 2024), with HS annex. https://www.meti.go.jp/policy/external_economy/trade_control/01_seido/04_seisai/downloadCrimea/20240222oshirase_russia.pdf
[^15]: METI, CHPI list overview (published 20 Oct 2023; updated 22 Feb 2024). https://www.meti.go.jp/policy/external_economy/trade_control/01_seido/04_seisai/downloadCrimea/20241023_gaiyo_russia.pdf
[^16]: METI-hosted G7 guidance (provisional JP translation) on preventing Russia export-control / sanctions evasion (CHPL, red flags, best practices). https://www.meti.go.jp/policy/external_economy/trade_control/01_seido/04_seisai/downloadCrimea/20241023_russia_guidance.pdf