Indonesia, AI governance tracker
Monsoon Ledger
AI governance from the archipelago, promises checked against the record.
Monsoon Ledger is a public, bilingual tracker of how Indonesia governs artificial intelligence. Every entry is tied to primary or official sources, labelled with how strong the evidence is, and compared with the EU AI Act and frontier AI safety practice. Key quotes stay in the original Bahasa Indonesia, with an English line under them.
The monsoon is forecast and announced long before it arrives. Indonesia's AI rules have worked the same way. This ledger records what was promised and what actually showed up.
01
Safety lens matrix
The same four questions asked of every entry. Select a cell to see the answer and the evidence behind it. Unclear means public sources do not answer the question yet. It does not mean no.
| Entry | Incident reporting | Unacceptable uses | General-purpose models | Enforcement |
|---|---|---|---|---|
| The AI rules that keep almost arriving | ||||
| Three risk tiers, not four | ||||
| A 72-hour breach clock, and nobody to call | ||||
| Report everything, not just high risk | ||||
| ASEAN's generative AI guide asks the region to explore incident reporting | ||||
| Circular Letter 9/2023 sets AI ethics guidance, with no incident duty | ||||
| Covered | 3 of 6 | 2 of 6 | 1 of 6 | 1 of 6 |
Select a cell to see the question, the answer and the evidence behind it.
Incident reporting
Does it require incidents to be reported?
Unclear
- Status
- Draft
- Evidence
- Official statementSaid on the record by an official. Shows intent, not law.
Unacceptable uses
Does it name uses that are not acceptable?
Unclear
- Status
- Draft
- Evidence
- Official statementSaid on the record by an official. Shows intent, not law.
General-purpose models
Does it address general-purpose models?
Unclear
- Status
- Draft
- Evidence
- Official statementSaid on the record by an official. Shows intent, not law.
Enforcement
How strongly is it enforced?
Unclear
- Status
- Draft
- Evidence
- Official statementSaid on the record by an official. Shows intent, not law.
Incident reporting
Does it require incidents to be reported?
Yes
- Status
- Draft
- Evidence
- Official statementSaid on the record by an official. Shows intent, not law.
Unacceptable uses
Does it name uses that are not acceptable?
Yes
- Status
- Draft
- Evidence
- Official statementSaid on the record by an official. Shows intent, not law.
General-purpose models
Does it address general-purpose models?
Unclear
- Status
- Draft
- Evidence
- Official statementSaid on the record by an official. Shows intent, not law.
Enforcement
How strongly is it enforced?
Soft
- Status
- Draft
- Evidence
- Official statementSaid on the record by an official. Shows intent, not law.
Incident reporting
Does it require incidents to be reported?
Yes
- Status
- Signed
- Evidence
- Press reportFrom journalism or secondary analysis. Provisional.
- Note
- Incident reporting here means personal data breaches, not AI incidents.
Unacceptable uses
Does it name uses that are not acceptable?
Unclear
- Status
- Signed
- Evidence
- Press reportFrom journalism or secondary analysis. Provisional.
- Note
- Incident reporting here means personal data breaches, not AI incidents.
General-purpose models
Does it address general-purpose models?
Unclear
- Status
- Signed
- Evidence
- Press reportFrom journalism or secondary analysis. Provisional.
- Note
- Incident reporting here means personal data breaches, not AI incidents.
Enforcement
How strongly is it enforced?
Unclear
- Status
- Signed
- Evidence
- Press reportFrom journalism or secondary analysis. Provisional.
- Note
- Incident reporting here means personal data breaches, not AI incidents.
Incident reporting
Does it require incidents to be reported?
Yes
- Status
- Draft
- Evidence
- Official statementSaid on the record by an official. Shows intent, not law.
Unacceptable uses
Does it name uses that are not acceptable?
Unclear
- Status
- Draft
- Evidence
- Official statementSaid on the record by an official. Shows intent, not law.
General-purpose models
Does it address general-purpose models?
Unclear
- Status
- Draft
- Evidence
- Official statementSaid on the record by an official. Shows intent, not law.
Enforcement
How strongly is it enforced?
Unclear
- Status
- Draft
- Evidence
- Official statementSaid on the record by an official. Shows intent, not law.
Incident reporting
Does it require incidents to be reported?
Unclear
- Status
- In force
- Evidence
- Official statementSaid on the record by an official. Shows intent, not law.
Unacceptable uses
Does it name uses that are not acceptable?
Unclear
- Status
- In force
- Evidence
- Official statementSaid on the record by an official. Shows intent, not law.
General-purpose models
Does it address general-purpose models?
Yes
- Status
- In force
- Evidence
- Official statementSaid on the record by an official. Shows intent, not law.
Enforcement
How strongly is it enforced?
None
- Status
- In force
- Evidence
- Official statementSaid on the record by an official. Shows intent, not law.
Incident reporting
Does it require incidents to be reported?
No
- Status
- In force
- Evidence
- Legal textRead in the signed or published text itself.
Unacceptable uses
Does it name uses that are not acceptable?
Yes
- Status
- In force
- Evidence
- Legal textRead in the signed or published text itself.
General-purpose models
Does it address general-purpose models?
No
- Status
- In force
- Evidence
- Legal textRead in the signed or published text itself.
Enforcement
How strongly is it enforced?
None
- Status
- In force
- Evidence
- Legal textRead in the signed or published text itself.
02
Promise vs Reality
Janji vs KenyataanThe public promises officials have made about Indonesia's AI rules, each with a computed outcome and a receipt.
4 of 5 dated promises missed as of
- 4Missed
- 1Open
- 0Met
Not counted: 1 conditional and 1 status claim, which set no deadline.
Most recent promises
- Task force starts once the President signsIrma Handayani, KomdigiNo deadline
- Only the President's signature leftEdwin Hidayat Abdullah, KomdigiNo deadline
- A priority to sign in 2026Meutya Hafid, MinisterOpen
03
Latest entries
The AI rules that keep almost arriving
Indonesia has two draft presidential regulations on AI, a national roadmap and an ethics rule. Officials have announced a finish line at least five times since January 2025, and both are still drafts.
DraftOfficial statementThree risk tiers, not four
The draft ethics regulation sorts AI into low, high and unacceptable risk, one fewer tier than the EU AI Act. High risk means a system uses specific personal data, unacceptable means it could threaten security and safety, and a public-service chatbot is the example of low risk. Each ministry sets its own risk standards for its sector. The rule carries no criminal sanctions, relies on self-assessment, and requires incident reporting channels.
DraftOfficial statementA 72-hour breach clock, and nobody to call
Government Regulation 33/2026, signed on 16 July 2026, sets out how Indonesia's 2022 personal data law works in practice. From 16 January 2027, organisations must notify affected people and the supervisory authority within 72 hours of confirming a breach, and the authority can fine up to 2% of annual gross revenue. But that authority has not been created. The presidential regulation needed to set it up has not been issued, and on 27 July 2026 the ministry opposed a court petition asking for a deadline.
SignedPress report
04
Radar
Twice a week a scanner checks Indonesian and ASEAN news and official legal databases. What it finds waits here, unreviewed, until the author reads the source.
Last scan: