AIR APACAI Readiness Index · Powered by Exeter Labs

JAPANQ2 2026

Japan publishes, and gets cited.

83 companies here reach the published APAC top 1,000.

Download the dossierPDF · 20 pages · 4.4 MB

Executive summary

Current, and widely cited

Readiness 58 of 100, range 52-64. Visibility 44.2 of 100 across 232 scored companies.

A low score here reflects the public record, not a company’s capability. Japan is among the record leaders: of the five signals the index reads, its record leans hardest on freshness and authority and thinnest on capability visibility. The conditions companies here operate under score 58, and what those companies publish scores 44.2. Of the two, the second is the one companies here control, and the quicker to change: 93% of the 232 companies with enough of a public trail to score publish no proof of capability at all.

Japan’s record is current and it is cited. The 17 companies that also publish proof of capability average 62 of 100, on a method the rest can copy.

Exhibit 1. Tier mix in this market
Tier mix in JapanEmerging 149, Findable 46, Building 27, Proven 10, of 232 listed companies in Japan. 2026-08-02T21:45:10.421033 image/svg+xml Matplotlib v3.11.1, https://matplotlib.org/

Source: AIR APAC AI Readiness Index, APAC 5000, Q2 2026. Scores measure public evidence, not verified capability.

83 companies in Japan reach the published APAC top 1,000.

149 of 232 are scored already, next in line for the published list.

58Readiness score for Japan. Range 52-64.
44.2Average visibility score across 232 scored companies. Public evidence only.
83Companies here that reach the published APAC top 1,000, of 232 scored.

Why visibility is the asset

Invisible is not the same as absent

Buyers, investors and partners treat it that way. Most decisions about a company are shaped before anyone is contacted, from what they can find. The record is the only version of it most people ever meet. and here most of the market already leaves one, which moves the question from whether a record exists to what it says about capability. In Japan, 103 of 346 companies screened leave too little behind to assess, which is the addressable group rather than the verdict.

Capital and customers screen before they call

Investors, corporate buyers and procurement teams build a shortlist from public material long before a meeting exists. A company absent from that material is not rejected on the merits. It never reaches the shortlist.

Partners and talent read the same record

Prospective partners, joint-venture counterparties and senior candidates all check what an organisation says about its own AI work. Silence is read as inactivity, which is often wrong and rarely without cost.

AI assistants now answer the question first

When someone asks an AI system what a company is doing with AI, the answer is assembled from public sources. A company that publishes nothing cedes its own description to whatever else the model finds.

Most companies we look at are doing more with AI than their customers, investors and partners can see. The work is real, the evidence of it is thin, and people decide on the evidence.
Indhran SegharFounding Director, AIR APAC
Proof concentration in Japan
Proof concentration in Japan17 of 232 ranked companies publish capability evidence, averaging 62 of 100. 93% publish none yet. 2026-08-02T21:45:10.477234 image/svg+xml Matplotlib v3.11.1, https://matplotlib.org/

Source: AIR APAC AI Readiness Index, APAC 5000, Q2 2026. Scores measure public evidence, not verified capability.

Evidence quality mix in Japan
Evidence quality mix in JapanShare of ranked companies in this market on each evidence grade A to D. 2026-08-02T21:45:10.503675 image/svg+xml Matplotlib v3.11.1, https://matplotlib.org/

Source: AIR APAC AI Readiness Index, APAC 5000, Q2 2026. Scores measure public evidence, not verified capability.

83Companies here reach the published APAC top 1,000.
71%Of 232 scored rest on grade A or B evidence.
149Scored already, and next in line for the published list.

Where the signal is strong

Japan’s strongest and weakest public signals

The five signals are not on one scale. Nearly every company in the index is discoverable and nearly none publishes proof of what it can do, so the longest bar below is a property of the index before it is a property of Japan. Read against its own other four signals, this record leans hardest on freshness and authority and thinnest on capability visibility. The first is where companies here have already shown they can publish; the second is where the same effort has not yet gone, and it is the clearest place to start.

Exhibit 2. Average score by signal in this market
Average score by signal in JapanDiscoverability: 62.3 of 100. AI & Digital Evidence: 48.1 of 100. Platform & Data Footprint: 36.6 of 100. Freshness & Authority: 60.9 of 100. Capability Visibility: 4.6 of 100. 2026-08-02T21:45:10.433531 image/svg+xml Matplotlib v3.11.1, https://matplotlib.org/

Source: AIR APAC AI Readiness Index, APAC 5000, Q2 2026. Scores measure public evidence, not verified capability.

Signal shape, Japan’s five signals
Signal shape, Japan's five dimensionsDiscoverability: 62 of 100. AI & Digital Evidence: 48 of 100. Platform & Data Footprint: 37 of 100. Freshness & Authority: 61 of 100. Capability Visibility: 5 of 100. 2026-08-02T21:45:10.464261 image/svg+xml Matplotlib v3.11.1, https://matplotlib.org/

Source: AIR APAC AI Readiness Index, APAC 5000, Q2 2026. Scores measure public evidence, not verified capability.

The lean is freshness and authority. 17 companies in Japan publish capability evidence, averaging 62 and well clear of the market. It works where it exists. The other 93% have yet to publish it, against 83% regionally.

Work published only in Japan’s own language is understated here, not absent.

Four tiers of visibility

Four tiers, and where companies in Japan sit

A tier is a position in the published regional ranking, not a grade anyone awards. It says how much of a company’s AI work a buyer or an investor can verify, and nothing at all about whether that work is any good.

TierJapanWhat it means
Tier 1Proven10Region rank 1-100. Publishing enough, recently enough, and across enough of the five signals that the public record alone establishes an AI position.
Tier 2Building27Region rank 101-400. A consistent public trail, usually strong on two or three signals and thin on evidence of hiring.
Tier 3Findable46Region rank 401-1,000. Credibly active and possible to verify, but sparse enough that it takes a deliberate search to find.
Tier 4Emerging149Ranked beyond 1,000. Clears the evidence threshold but does not reach the published list. Real companies, scored and unlisted, and the largest group in almost every market.

10 of the 232 companies in Japan that clear the threshold reach tier 1, and 149 are emerging, 64% of those scored. Emerging is the normal state of the region, and the group standing closest to the tier above.

Moving up a tier is a disclosure decision, not a technology one, because every signal this index reads is something a company already controls.

The visibility ladder in Japan
The visibility ladder in JapanScreened 346, scored 243, listed 232, published 83. 2026-08-02T21:45:10.513675 image/svg+xml Matplotlib v3.11.1, https://matplotlib.org/

Source: AIR APAC AI Readiness Index, APAC 5000, Q2 2026. Scores measure public evidence, not verified capability.

Sector strengths in Japan
Sector strengths in JapanManufacturing: 47 companies, mean 47.7. Tech/Fintech: 41 companies, mean 45.6. Professional Services: 40 companies, mean 44.6. Financial Services: 47 companies, mean 44.4. Logistics/Supply Chain: 29 companies, mean 42.5. 2026-08-02T21:45:10.526425 image/svg+xml Matplotlib v3.11.1, https://matplotlib.org/

Source: AIR APAC AI Readiness Index, APAC 5000, Q2 2026. Scores measure public evidence, not verified capability.

National champions in Japan, top ten by visibility
National champions in Japan, top ten by visibilityLegalOn Technologies 71. Renesas Electronics Corporation 71. MUFG Bank, Ltd. 68. ITOCHU Techno-Solutions Corporation 66. Mercari 66. Mitsubishi Electric Corporation 66. NS Solutions Corporation 66. Preferred Networks, Inc. 66. SCSK Corporation 66. Mitsubishi Research Institute 65. 2026-08-02T21:45:10.548266 image/svg+xml Matplotlib v3.11.1, https://matplotlib.org/

Source: AIR APAC AI Readiness Index, APAC 5000, Q2 2026. Scores measure public evidence, not verified capability.

58CERS (0 to 100)triangulated market signal
52-64CERS band
BConfidence grade
221Domestic signal entities
44.2Mean raw RPScompany visibility

Source: AIR APAC AI Readiness Pulse, Q2 2026. CERS is a market-level signal, not a company readiness score. View sources.

Market context

Macro, state, and enterprise benchmarks for Japan.

BenchmarkValueRole
Country Readiness Prior (CRP v0.1)74.7Macro and state readiness prior
CRP v0.2 (three-block)75.5Full triangulated prior
Oxford Government AI Readiness 202475.8Macro input to CRP
IMF AI Preparedness Index 202373.3Macro input to CRP
Enterprise adoption (Salesforce 2023)42.1%Business AI adoption block
Exhibit 3. Japan's readiness score and published range
Japan's readiness score and published rangeJapan 58, range 52-64. 2026-08-02T21:45:10.452715 image/svg+xml Matplotlib v3.11.1, https://matplotlib.org/

Source: AIR APAC AI Readiness Index, APAC 5000, Q2 2026. Scores measure public evidence, not verified capability.

Japan carries the enterprise-adoption block. CERS triangulates the macro prior, public company signal, and enterprise adoption with a coverage-based shrinkage; the precise integer is indicative, so a band is also published.

Public company signal

What the indexed companies in Japan show.

SignalValue
Mean raw RPS visibility44.2
Sector-stratified signal43.7
Supplier-readiness signal44.2
Domestic-adopter signal46.1

Top organisations

The highest public-signal organisations in Japan.

RankOrganisationSectorRPSEvidence
22LegalOn TechnologiesTech/Fintech71.0B
24Renesas Electronics CorporationManufacturing71.0B
53MUFG Bank, Ltd.Financial Services68.0A
68ITOCHU Techno-Solutions CorporationProfessional Services66.0B
72MercariRetail/E-commerce66.0B
73Mitsubishi Electric CorporationManufacturing66.0B
74NS Solutions CorporationProfessional Services66.0B
75Preferred Networks, Inc.Tech/Fintech66.0B
78SCSK CorporationProfessional Services66.0B
99Mitsubishi Research InstituteProfessional Services65.0A
101NEC CorporationTech/Fintech65.0B
107PKSHA Technology Inc.Professional Services65.0A
134Yokogawa Electric CorporationManufacturing64.0B
142Denso CorporationManufacturing63.0B
152Sompo Holdings, Inc.Financial Services63.0A
158Amazon Japan G.K.Retail/E-commerce62.0C
184JFE HoldingsManufacturing62.0A
193Nissan MotorManufacturing62.0A
194Omron CorporationManufacturing62.0B
195ORIX CorporationFinancial Services62.0B

Ranked by raw public signal (RPS), a visibility measure. Open any organisation for its evidence profile.

Limits and what this is not

What this report does not claim

Printed with the findings rather than in an appendix at the back.

Known blind spots in this market

Japan carries 2 this cycle: evidence is gathered from English-language search, likely undercounting local-language activity; below the full three-source-block coverage standard. Each is a direction we know the number leans in. The country evidence behind the readiness score is graded B on source coverage of macro=1, enterprise=1.

A high score means published, not proven

A high visibility score means a company publishes a lot that can be found about its AI work. It does not mean the work is good, in production, or making money. A company here running serious AI quietly will score low here, and it should: this index reads the public record and nothing else.

What would change these numbers

Three things, all scheduled and all documented on page 21. Fuller source coverage per market, which is what currently sets the width of every readiness range. Validating the coverage weighting behind those ranges, which is a stopgap in this edition. And merging companies that operate several web domains, so one group cannot appear twice. Nothing here was audited inside a company, and no verified-readiness score is published this cycle. Next edition: Q3 2026.

english_web_undercoveragebelow_full_crp_v02_coverage

What to do with a visibility score

A visibility score is something a company here can move, because every signal behind it is something it already controls. Three steps separate it from a record that shows what it is actually doing.

Start with the company’s own row

Every scored company in Japan appears in the dataset with its five signal scores and an evidence grade. A grade C or D row is usually a company that has not said much, not a company that has not done much.

Then measure what this cannot see

This index reads the outside only. Leadership alignment, governance, data quality and workforce capability are invisible to it by design, and that is where the real answer sits.

Correct the record on purpose

Proof of capability is the region’s weakest signal, because companies announce intent faster than they show work. Job posts, named leads and dated case studies move the score honestly.

None of this requires a new AI programme. It requires publishing the one that already exists.

The full Japan dossier, as published.

83 of the 232 ranked companies in Japan reach the published APAC top 1,000.

Macau has no benchmark block and is reported as contextual. The dossier counts companies on a wider definition than this page does and the two reconcile: how the company counts fit together.

Assess readiness