Role
AI product manager
Shipping the system, and holding the duties that fall on whoever puts it on the market.
§ 1 — What you carry
Where the duty attaches
The provider carries the heaviest column in the corpus: the data governance, the technical documentation, the record-keeping, the post-market monitoring. These are not release-gate checks. Most of them are continuing duties whose evidence has to exist for as long as the system is on the market.
The status is easier to acquire than teams expect. Rebranding a bought system, or modifying one substantially, makes a provider of the organisation doing it — with the whole column attached, and usually with none of the documentation the original provider holds.
Also written as product owner, head of product, AI programme lead. Everything below is drawn from the obligation dataset by the capacity that binds this role, not by job title — 4 provisions across 1 collection.
§ 3 — Evidence
The artefacts this role owns
Drawn from the evidence column of the provisions above, most widely demanded first. These are what an assessor asks this role to produce.
- 1xA bias examination that reached the risk register
- 1xA dataset defect register, with treatment
- 1xA dated residual-risk acceptance decision
- 1xA recorded obviousness assessment for 50(1)
- 1xA risk management plan tied to a named system and release
- 1xCybersecurity measures addressing AI-specific attack surfaces
- 1xDeclared accuracy levels and metrics in the instructions for use
- 1xEvidence that machine-readable marking actually survives
- 1xEvidence that post-market data re-entered the process
- 1xFeedback-loop mitigation records for systems that continue to learn
- 1xFor edited text, both limbs of the exception
- 1xProvenance and original collection purpose for personal data
§ 4 — Ahead
What lands next
2 December 2026
Article 50 — Transparency obligations for providers and deployers of certain AI systems
2 December 2027
2 December 2027
2 December 2027
2 August 2028
2 August 2028
2 August 2028
§ 5 — Examined in
Where this is examined
Read from the syllabus entries the provisions above actually carry, not matched by job title. The count is how many of this role's obligations each credential examines, so you can see how much of the role it covers.
- Certified AI Compliance Fundamentals — EU AI Act
Examines 4 of the 4 provisions above, under General-purpose AI and transparency, High-risk obligations and conformity.
- Certified AI Audit Fundamentals
Examines 2 of the 4 provisions above, under Control identification and testing, Evidence and working papers.
- Agentic AI Oversight Endorsement
Examines 1 of the 4 provisions above, under End-user responsibility and multi-agent risk.
- Certified AI Compliance Fundamentals — Data Protection
Examines 1 of the 4 provisions above, under Data-protection foundations applied to AI.
- Certified AI Recruitment Bias — Fundamentals
Examines 1 of the 4 provisions above, under Disparate-impact fundamentals.