In brief
On 20 May 2026, the Infocomm Media Development Authority (IMDA) updated the Model AI Governance Framework for Agentic AI (MGF). The MGF, first rolled out in January 2026, provides guidance on managing the risks associated with deploying agentic AI systems which are capable of autonomous planning, reasoning, and action.
The updated MGF incorporates industry feedback and introduces new best practices and real-world case studies, demonstrating how organisations have operationalised the framework to address multi-agent systems, third-party agents, automation bias, amongst other recommendations.
In more detail
The MGF provides guidance for organisations to reduce agentic AI deployment risks, which are structured around four core dimensions: (1) assessing and bounding the risks, (2) ensuring meaningful human accountability, (3) implementing technical controls and processes, and (4) enabling end-user responsibility. You may find out more about the MGF from our previous client alert here.
Key updates to the MGF include the following:
- Components of agentic AI: The addition of safety and reliability components, including access controls, guardrails, human approvals, logging and monitoring, as core components of an AI agent.
- Risks of agentic AI: The inclusion of risks arising from multi-agent systems and the addition of third-party agent usage and agentic system complexity as factors affecting the likelihood of risk materialising.
- Meaningful human accountability: A differentiation of the roles and responsibilities of platform providers versus system providers or app developers across the agentic AI value chain and lifecycle, encouraging adaptive governance and providing further guidance on guarding against automation bias (e.g., monitoring human override rates and response times).
- Technical controls and processes: Guidance on selecting appropriate technical controls for each risk category at design, development, pre-deployment and deployment stages (e.g., structural, rule-based versus model-based or prompt-layer controls), and recommendations for robust change management processes to prevent small modifications from cascading into larger impacts as system complexity increases.
- End-user responsibility: An elaboration of the potential impact on tradecraft and business continuity if agents take over entry-level tasks, such as skill degradation and loss of basic operational knowledge to perform critical processes manually when agents malfunction or become unavailable.
The updated MGF also includes case studies across all four MGF dimensions, contributed by Singaporean companies, multinational enterprises and government agencies. The case studies serve as practical and actionable illustrations of how organisations have operationalised the MGF's recommendations in real-world agentic AI deployments across diverse industries.
Key takeaways
The latest updates to the MGF provide clearer practical guidance for organisations deploying agentic AI, particularly in relation to multi-agent systems. Organisations deploying or considering the deployment of agentic AI systems should review the updated MGF and assess the alignment of their internal governance controls and processes with the MGF's recommendations.
For further information and to discuss what this development might mean for you, please get in touch with your usual Baker McKenzie contact.
* * * * *
© 2026 Baker & McKenzie. Wong & Leow. All rights reserved. Baker & McKenzie. Wong & Leow is incorporated with limited liability and is a member firm of Baker & McKenzie International, a global law firm with member law firms around the world. In accordance with the common terminology used in professional service organizations, reference to a "principal" means a person who is a partner, or equivalent, in such a law firm. Similarly, reference to an "office" means an office of any such law firm. This may qualify as "Attorney Advertising" requiring notice in some jurisdictions. Prior results do not guarantee a similar outcome.