20 July 2026
In brief
On 9 June 2026, the Department of Information and Communications Technology (DICT) and the Civil Service Commission (CSC) issued Joint Memorandum Circular No. 003, Series of 2026 (JMC), setting out principles for the ethical and responsible development, deployment, and use of artificial intelligence (AI) in government. The JMC applies broadly across public sector entities and establishes expectations around governance, accountability, and data protection. While directed at government agencies, the framework is also relevant to private sector organizations that develop, supply, or support AI systems for public sector use.
Key takeaways
- Government agencies must align their AI-related activities with a set of principles covering fairness, transparency, accountability, privacy, and security.
- All covered entities are required to establish oversight mechanisms, including internal committees, for AI governance.
- AI developers, deployers, users, and vendors are assigned specific responsibilities, including accountability for outcomes and safeguards against bias and misuse.
- Private sector vendors may face downstream requirements, as government agencies remain accountable for outsourced AI systems and must ensure third-party compliance.
- Organizations engaging with the public sector should consider reviewing governance frameworks, contractual arrangements, and AI lifecycle controls in light of the JMC.
In more detail
The JMC establishes a principles-based framework for the ethical and responsible use of AI within the Philippine government, framed against constitutional mandates, the Data Privacy Act of 2012, and international AI governance standards. It seeks to promote accountability, human rights protection, and public trust while enabling the responsible adoption of AI technologies in public service. In particular, the JMC sets out the scope of application, core principles, roles and responsibilities, and governance and compliance mechanisms for AI systems in the public sector.
Scope of application
The JMC applies to all government officials and employees across national agencies, constitutional bodies, government-owned or controlled corporations, local government units, and state universities and colleges. It also extends to individuals engaged under job orders or service contracts whose work involves AI systems. Agencies are required to ensure that such individuals comply with the JMC, indicating that requirements may extend to outsourced and third-party engagements.
Definition of AI systems and stakeholders
The JMC adopts a broad definition of AI systems, covering machine learning, generative AI, autonomous systems, and similar technologies that generate outputs such as recommendations or decisions. It identifies key stakeholders—AI developers, deployers, users, and vendors—and assigns responsibilities to each. Notably, the definition of AI developer and deployer includes situations where development or deployment is outsourced, with the procuring agency retaining accountability.
Core AI principles
The JMC adopts a set of guiding principles drawn from international frameworks (OECD, UNESCO, and ASEAN), including:
- Proportionality and Do No Harm: AI systems must not cause or exacerbate harm, with safeguards and risk assessments required where risks exist, and strict prohibition on uses that infringe privacy or human rights.
- Human-Centricity: AI must serve human well-being, with human oversight required in high-risk decisions and avoidance of manipulative or deceptive practices.
- Fairness and Non-Discrimination: AI systems must prevent bias and discriminatory outcomes, with mechanisms such as bias audits and mitigation strategies.
- Transparency and Explainability: Agencies must ensure that AI processes and outputs are understandable, with clear disclosures regarding purpose, inputs, risks, and limitations.
- Data Governance, Privacy, and Protection: AI systems must incorporate safeguards aligned with the Data Privacy Act, including privacy impact assessments.
- Responsibility and Accountability: Developers and deployers are accountable for AI outcomes, including where development is outsourced.
- Other Principles: These include multistakeholder governance, cybersecurity, robustness and reliability, and sustainability considerations.
Roles and responsibilities
The JMC assigns responsibilities to different stakeholders:
- AI developers must ensure responsible design, testing, and documentation of AI systems.
- AI developers must monitor system performance, ensure data integrity, implement governance mechanisms, and provide disclosures regarding AI use.
- AI users remain accountable for their use of AI and may challenge AI-driven decisions, with a right to human review and redress.
- AI vendors must deliver reliable and secure AI solutions and disclose risks and limitations.
Governance, oversight, and compliance
Government agencies are required to establish oversight and review mechanisms, including an internal committee composed of representatives from IT, legal, and relevant operational units. These committees are responsible for guiding AI use and making recommendations to agency heads, who remain accountable for approval decisions.
The JMC further requires agencies to adopt or update internal policies aligned with its principles, integrate AI governance into their Information Systems Strategic Plans and procurement processes, and monitor compliance on an ongoing basis.
Data handling and accountability measures
Agencies must classify and process government data used in AI systems in accordance with applicable laws, including the Data Privacy Act and other regulations on public information, security, and records management. The JMC emphasizes transparency, accountability, and human oversight, with violations subject to penalties under existing laws.
Effectivity and next steps
The JMC takes effect immediately upon compliance with publication requirements. The DICT and CSC are mandated to issue implementing guidelines within 90 days from effectivity to operationalize its provisions.
How this affects your business
Private sector organizations, particularly those developing, supplying, or deploying AI systems for government agencies in the Philippines, may face downstream requirements, as agencies are expected to ensure that third-party providers adhere to the same principles under the JMC. Businesses should therefore consider reviewing their AI governance, data protection, and risk management frameworks to assess alignment with the JMC’s requirements. Companies engaging with public sector clients may also wish to review contractual arrangements and internal processes to ensure appropriate documentation, disclosures, and safeguards are in place across the AI lifecycle.
For further information on how this development may affect your organization, please feel free to reach out to our team in Quisumbing Torres, and we will be pleased to assist.