Bridging the AI Readiness Gap in Asia Pacific
Mid-market enterprises, and why the tools built to measure them cannot see them. The inaugural AIR APAC white paper: 510 companies scored across five markets and six sectors from behavioural public signals.
This is the inaugural AIR APAC white paper (April 2026): Q1 2026 data on 510 mid-market companies in five markets. For current results, see the Q2 2026 AI Readiness Index.
Open the paper
The paper opens direct. No form, no account, no email address.
Across Asia Pacific, the 200 to 3,000 employee firms that form the backbone of every market are struggling to turn AI ambition into execution. The paper finds that the first problem is not capability but legibility: the platforms built to measure AI readiness were built for North American and European enterprises, and they undercount APAC companies in two ways, by leaving firms out entirely and by mis-measuring the ones they do list.
AIR APAC calls this the APAC Data Desert. It matters because a pipeline that returns no data cannot tell a company with no AI capability from a company that is simply not observable, so a cross-market score is a composite of readiness and observability. The paper sets out the method, the quadrant framework, the twelve-market ecosystem context, and the human capital constraint that survives every adjustment: no market in the ecosystem analysis scores above 65 on AI Talent Pool.
The basis for this paper
- Claim
- The APAC mid-market is not only lagging in AI adoption; it is systematically invisible to the tools used to measure it, and the mid-market is where the readiness gap is widest.
- Source
- Bridging the AI Readiness Gap in Asia Pacific, Mid-Market Enterprises. AIR APAC white paper, first edition, version 2, April 2026. This study came before the AI Readiness Index. Published by AIR APAC; the instruments and pipeline are built and operated by Exeter Labs.
- Period
- Q1 2026 data vintage. The paper was published in April 2026.
- Population
- 510 mid-market companies of 200 to 3,000 employees, scored across five markets (Singapore, Australia, Malaysia, Indonesia and Thailand) and six sectors, with a twelve-market ecosystem analysis for regional context.
- Method
- Behavioural public signals rather than self-reported surveys: cloud platform signals, specialist hiring activity, ERP system detection and analysed strategic communications. The company score is a public-evidence visibility score, Total Readiness Score against Data Visibility Score, not verified readiness.
- Limitation
- Most scores in this paper rest on thin public evidence and show a direction only. This is the 'APAC Data Desert' the paper describes. Cross-market scores combine readiness with observability. The twelve-market ecosystem scores are country-level context and are not part of the company index.
What this paper does not claim
A high score means the conditions for AI are visible in public, not that a company has deployed AI or earned a return from it. A company missing from the data is not shown to be inactive.
Findings in figures
Two figures carry the central finding. The first shows readiness falling as measurement quality falls; the second shows the same desert at country level, where the AI Talent dimension is the weakest cell in almost every market.
More research on organisations
More research on organisations in Asia Pacific.
- Asymmetric Visibility, Q1 2026 findings report: six findings from the same 510-company dataset, PDF, 43 pages, 0.6 MB.
- AI Readiness Index, Q2 2026 data: the published 1,000 organisations as a CSV, 116 KB, 1,000 rows.
- How the Q2 2026 edition is built: the current method on one page.
- What this index cannot tell you: the published limits of the Index.
Next step
Read the market patterns
The organisation-level evidence continues on the market pages: per-market readiness, the scored set, and the dossier for each. If you need a sector reading, the sector index sits beside it.