AI Readiness Checklist for Malaysian SMEs (20 Questions)
A 20-question AI readiness checklist for Malaysian SMEs covering goals, data, people, systems and risk, with scoring and what to do at each level.
You are ready for an AI project when you can name a process and a result, your data is accessible, one person owns the outcome, staff have time to test and you have a personal-data rule. Score the 20 questions below: 15 or more means build, 8 to 14 means prepare, under 8 means tidy and train first.
Part of the guide: AI Consulting in Malaysia: What It Costs and What It Delivers (2026)

On this page
Readiness matters more than the choice of tool. Surveys of Malaysian SMEs point to the same blockers: an Ecosystm study backed by Red Hat and the National AI Office found 60% of 133 surveyed SMEs named lack of in-house technical expertise and 52% named the cost of implementation and maintenance. This checklist turns those blockers into questions you can answer in ten minutes. For the wider picture, see 30 statistics on AI adoption in Malaysia.
How do you use the checklist?
Answer yes or no, one point per yes, and be strict. "Partly" is a no. Do it for one process, not for the whole company.

Goal and process
- Can you name one process you want to improve?
- Can you state the result in hours, ringgit or days (for example "cut quote time from two days to four hours")?
- Is the process repeated at least weekly?
- Do you know roughly how many hours it takes now?
- Would a wrong output be cheap to catch before it reaches a customer?
Data and systems
- Does the data for this process sit in a system or documents you can open today?
- Is it mostly digital, not paper or voice notes?
- Is it reasonably consistent (same fields, similar formats)?
- Do the key systems (CRM, accounting, ERP) have an export or API?
- Do you know which of this data is personal data?

People
- Is one named person responsible for the outcome?
- Will the people who do the work be involved in the design?
- Do staff have time (about one hour a week) to test during the pilot?
- Has anyone on the team used an AI assistant on real work?
- Is leadership willing to change the process, not only add a tool?
Risk and governance
- Do you have a written rule on what may be entered into AI tools?
- Do you know where each tool stores data and who can see it?
- Is there a review step for anything that reaches a customer?
- Can you stop or switch off the system quickly if it misbehaves?
- Do you know whether a data processing agreement is needed with the vendor?
For questions 10, 16, 17 and 20 see PDPA and AI integration in Malaysia.
What does your score mean?
| Score | Level | Next step |
|---|---|---|
| 15 to 20 | Ready | Scope a build. Use the 90-day plan |
| 8 to 14 | Close | Fix the gaps, run a small pilot with ready-made tools, train the team |
| 0 to 7 | Not yet | Tidy the data, write the policy, train one person, then retry |
Read the pattern as well as the total. A score driven by low numbers in the data section means cleanup is the project. Low numbers in people mean training and ownership come before technology; in-house sessions start from MYR 4,500 and can be HRD Corp Claimable, see the HRD Corp guide.
How does this relate to the winsym.ai audit?
Our AI Opportunity Audit scores readiness as a WTX value from 0 to 100 across seven dimensions, based on interviews and document samples, not self-assessment. This checklist is the free first pass. Self-scores tend to run high, so compare your answers with a colleague's.
What should you do next?
Fill it in today for your best candidate process, then book a free 30-minute call and bring the score. We will say what to do first. For the larger picture, see AI consulting in Malaysia: costs and deliverables, and for choosing the process, pick your first AI use case.
Sources
Links checked November 2026.
- Red Hat, Red Hat and NAIO study of 133 Malaysian SMEs, press release, 30 October 2025 (study by Ecosystm; 36% piloting, 21% scaled beyond pilots, 60% cite lack of in-house technical expertise).
- winsym.ai AI Opportunity Audit and WTX score, winsym-ai.my/audit, checked 22 October 2026.
- Personal Data Protection Department, Personal Data Protection Act 2010.
Not an official standard. A working tool built from our experience.
Frequently asked questions
What does AI readiness mean?
It means the conditions that decide whether an AI project reaches daily use: a clear goal, accessible data, an owner, staff time and a data-handling rule. Technology is rarely the limit.
What score do I need to start?
Fifteen or more of twenty is a good basis for a custom build. Between eight and fourteen, close the gaps first or start with training and a small pilot. Below eight, begin with data tidying and a policy.
Is this checklist an official standard?
No. It is the winsym.ai working checklist, based on our audits and on the common causes of stalled pilots. It is not a government or ISO assessment.
Who should fill it in?
The owner and the manager of the process you want to improve, separately. Differences in their answers are often the most useful finding.