PitchMyStory Newswire
Malaysia Is Building Rules for Trustworthy AI - But What Information Should AI Be Allowed to Trust?
As Malaysia moves toward a national AI governance framework, Kuala Lumpur-developed Proof of Knowledge argues that responsible AI requires organisations to know whether the information their systems use is approved, current and verifiable.
KUALA LUMPUR, MALAYSIA | 8 September 2026
Malaysia is putting new institutions, standards and policy frameworks in place to support responsible artificial intelligence. The Ministry of Digital began engagement in July on a proposed Artificial Intelligence Governance Bill, describing it as Malaysia's first horizontal legal framework dedicated specifically to AI governance.
The proposed framework uses a risk-based approach. Malaysia has also launched AI Malaysia Berhad, introduced the National AI Action Plan 2026-2030 and established the Malaysia AI Safety Institute. Together, these measures support the country's ambition to become an AI Nation by 2030 while strengthening governance, safety and public confidence.
Kuala Lumpur-developed technology venture Proof of Knowledge (POK) believes this direction raises a practical question for every organisation adopting AI: what information should an AI system be allowed to rely upon?
When AI relies on the wrong information
An AI system can be secure, well governed and operating as designed, yet still produce an unsafe or misleading answer if its source material is outdated, unapproved or impossible to verify. The risk becomes greater when an answer informs a decision, generates a report, recommends an action or enters an automated workflow.
Consider an employee asking an AI assistant for the current procedure for responding to a safety incident. The system may find three similar documents: an approved procedure that has been replaced, a working draft and the current approved version. All three may look legitimate. Only one should guide action.
The same problem can arise with compliance policies, engineering instructions, government circulars, contracts and professional guidance. Information is frequently copied, downloaded, emailed, summarised or moved between systems. In that process, the connection between a document and the evidence of its origin, approval and status can weaken.
AI governance and information governance
“Governing the AI model is essential, but it does not answer every question,” said Jordan Richards, co-founder of Proof of Knowledge. “Organisations also need to know what information the system used, where it came from, whether it was approved and whether a newer version has replaced it.”
AI governance sets responsibilities for how AI is developed, deployed, monitored and controlled. Information governance deals with the records, policies and knowledge on which people and systems rely. The two disciplines overlap, but they are not the same.