Artificial Intelligence Regulation Act (Aira)

TWENTIETH CONGRESS OF THE REPUBLIC OF THE PHILIPPINES ) First Regular Session 25 JUL -2 P4:55 SENATE S. No. 25 p... Introduced by Senator PIA S. CAYETANO AN ACT REGULATING THE DEVELOPMENT AND USE OF ARTIFICIAL INTELLIGENCE SYSTEMS IN THE PHILIPPINES, PROMOTING ETHICAL AND RESPONSIBLE ARTIFICIAL INTELLIGENCE INNOVATION, AND INTEGRATING SUSTAINABILITY AND FUTURES THINKING IN NATIONAL POLICY MAKING, AND FOR OTHER PURPOSES EXPLANATORY NOTE The rise of Artificial Intelligence (AI) is profoundly transforming industries, governance systems, and societies worldwide. As the Philippines continues its digital transformation journey, there is an urgent need for a national framework that ensures the safe, responsible, and ethical use of AI, aligned with our vision of an inclusive, innovative, and secure digital future. The European Union's landmark AI Act of 2024, the world's first comprehensive legal framework on AI, highlights the importance of risk-based classification, transparency, accountability, and human oversight.' In the same way, the United Nations Secretary-General's High-Level Advisory Body on Al emphasizes the need for coordinated global governance to uphold rights-based, transparent, and inclusive Al development.2 These international efforts reflect a growing global recognition. While AI holds vast economic and societal potential, it must be accompanied by a robust regulatory legal framework. ' European Union AI Act. (2024). World's First Comprehensive AT Law https://digital-strategy.ec.europa.eu/en/policies/european-approach-artificial-intelligence 2 CDO Magazine. (2024). UN advisory body's 7 key recommendations for global Al governance. https://www.cdomagazine.tech/aiml/un-advisory-bodys-7-key-recommendations-for-global-ai- governance?utm

Al presents enormous opportunities for the Philippines: improving public services, advancing disaster resilience, modernizing agriculture, and enhancing education and healthcare systems, among others. However, these innovations carry risks. AI systems rely on data and models that may be incomplete, biased, or manipulated, reflecting the subjective or commercial choices of developers. These raise concerns such as algorithmic bias, discrimination, and AI hallucinations, where systems generate false or misleading outputs with confidence. Moreover, the global scientific and policy communities have raised alarms regarding the possible rise of Artificial Superintelligence (ASI), hypothetical Al systems surpassing human intelligence and potentially beyond human control.3 This risk includes fears that such systems, if improperly secured or regulated, might gain unauthorized access to critical infrastructure, including military assets such as nuclear weapons, posing existential threats to humanity. In view of these challenges, this bill seeks to strike a careful balance between encouraging technological innovation and ensuring that Al systems remain safe, ethical, transparent, and under meaningful human oversight. Given that Al is still in its early stages, this bill provides a general framework to encourage its development in a responsible and lawful manner. It envisions a future where Al supports Filipino ingenuity, addresses national development challenges, and protects the rights and welfare of every citizen. The State bears the responsibility to ensure that AI will never be used to perpetrate crimes, abuse rights, cause harm or unintended consequences, whether through intent or accident, while supporting Filipino ingenuity and technological progress. In view of the foregoing, the approval of this bill is earnestly sought. Pis P. Centro 3 Center for Security and Emerging Technology. (2024). Securing Critical Infrastructure in the Age of Al. Georgetown University https://cset.georgetown.edu/publication/securing-critical-infrastructure-in-the- age-of-ai/

TWENTIETH CONGRESS OF THE REPUBLIC OF THE PHILIPPINES First Regular Session 25 JUL -2 P4:56 SENATE RECT S. No. _ Introduced by Senator PIA S. CAYETANO AN ACT REGULATING THE DEVELOPMENT AND USE OF ARTIFICIAL INTELLIGENCE SYSTEMS IN THE PHILIPPINES, PROMOTING ETHICAL AND RESPONSIBLE ARTIFICIAL INTELLIGENCE AND INTEGRATING INNOVATION, SUSTAINABILITY AND FUTURES THINKING IN NATIONAL POLICY MAKING, AND FOR OTHER PURPOSES Be it enacted by the Senate and House of Representatives of the Philippines in Congress assembled:

Section 1. Short Title. - This Act shall be known as the "Artificial Intelligence

Regulation Act (AIRA)."

Sec. 2. Declaration of Policy. - It is the policy of the State to recognize that

4 science and technology are essential for national development and progress. The State 5 shall give priority to research and development, invention, innovation, and their 6 utilization; and to science and technology education, training, and services.' In this regard, the State shall promote the responsible development and use of Artificial Intelligence (AI) to advance inclusive growth, public service delivery, innovation, and long-term national resilience. It shall likewise institutionalize futures thinking and sustainability as core principles in education, governance, and innovation. Towards this end, the State shall advance Al in a manner that is ethical, inclusive, transparent, and accountable.

Sec. 3. Objectives. - The objectives of this Act are the following:

^ Article XIV, Section 10 of the Philippine Constitution

a) Promote innovation, technological advancement, and the responsible and ethical use of AI in a manner that upholds the privacy, rights, safety, and dignity of all Filipinos; b) Recognize that Al enhances human productivity, creativity, and efficiency across all sectors, and may be harnessed as a tool to complement human work, improve public service delivery, boost economic competitiveness, and advance scientific, educational, and technological development; c) Ensure that AI systems should enhance human cognition and never degrade it; d) Strike a balance between promoting technological advancements on AI and ensuring Al safety for purposes of public interest; e) Protect Filipino workers from undue displacement due to AI and to uphold their right to decent and sustainable work; f) Ensure that AI systems contribute to inclusive national development that protects national security and respects human autonomy; g) Adapt to the rapid and evolving nature of AI technologies and continuously build institutional and regulatory capacities; and h) Ensure that policies of both the national and local governments on Al are driven by futures thinking, strategic foresight, and proactive risk management, and are governed by principles of transparency, accountability, and sustainability.

Sec. 4. Coverage. - This Act shall regulate the development and use of all types

of Al, including Artificial General Intelligence (AGI) and Artificial Superintelligence (ASI), as well as regulate Al foundation models, such as machine learning systems, generative AI, neural networks, expert systems, language learned models (LLMs), and Generative Pre-trained Transformers. This shall apply to all individuals, corporations, institutions, and government agencies that develop, deploy, use or operate AI systems in the Philippines.

Sec. 5. Definition of Terms. - For the purposes of this Act:

a) Artificial Intelligence (AI) - refers to systems that allow machines to think like humans, such that they can display intelligent behavior by

analyzing the data and taking actions with some degree of autonomy to achieve specific goals. b) Artificial General Intelligence (AGI) - refers to AI systems with the capacity to understand, learn, and apply knowledge across a broad range of tasks at a level equal to or surpassing human intelligence. c) Artificial Superintelligence (AST) - refers to hypothetical AI systems that surpass human intelligence in all respects, including creativity, decision- making, and social intelligence. d) Al Foundation Model - refers to a type of large-scale artificial intelligence model trained on vast quantities of broad data using self-supervised learning or similar techniques, and which can be adapted to a wide range of downstream tasks, such as question answering, summarization, translation, classification, or content generation, with minimal task- specific tuning. Foundation models include, but are not limited to, LLMs, multimodal models, and generative models. e) Deployment in Al - refers to the process of integrating and operating a trained AI model into an organizational infrastructure or real-world environment where it can perform its intended tasks. f) Expert Systems - refer to computer programs designed to simulate the decision-making abilities of a human expert in a specific domain. g) Futures Thinking - refers to an avenue to strategically explore a range of "possible futures", with the aim of uncovering unexpected opportunities and mitigate potential risks. h) Generative Pre-trained Transformers - refer to a type of large language model based on the transformer architecture that generate human-like text by predicting the next word in a sequence. i) Hallucination in Al - refers to the phenomenon wherein an Al system generates output that is syntactically correct or plausible-sounding but factually incorrect, fabricated, or nonsensical. Hallucinations can occur in text, image, audio, or multimodal outputs, and are often the result of the model extrapolating beyond the data it was trained on or failing to

ground its responses in verifiable information. There are two (2) general types: i) Intrinsic hallucination - when the model generates content inconsistent with the source input or prompt. ii) Extrinsic hallucination - when the content is not supported by the input but cannot be definitively labeled as incorrect due to lack of ground truth. j) Highest level of responsibility and ethical consideration - refers to the level of diligence required of all persons, corporations, institutions, and organizations engaged in the development and deployment of Al to ensure that Al systems are developed and used for the common good and does not facilitate crime, harm, or social injustice. k) Machine Learning (ML) - refers to a subset of AI that enables systems to learn from data and improve performance over time without being explicitly programmed. 1) Neural Networks - refer to computational models inspired by the structure and function of the human brain, designed to recognize patterns and relationships in data. m) Generative AI - refers to AI systems capable of producing text, images, music, video, or other content including LLM. n) Large Language Models (LLMs) - refer to a type of generative AI trained on vast datasets to generate human-like text, translate language, summarize information, or answer questions. 0) Multimodal Models- refer to systems designed to process and integrate information from multiple data modalities-such as text, images, audio, and video-within a single unified framework. These models learn to understand and generate responses by finding relationships and patterns across different types of inputs.

Sec. 5. National Al Commission. - There is hereby established a policy-making

and quasi judicial body to be known as the National Al Commission, hereinafter referred to as "NAIC." The NAIC shall be an agency attached to the Department of Science and Technology (DOST).

Sec. 6. Jurisdiction of the NAIC. - The NAIC shall have the original and exclusive

jurisdiction over all matters pertaining to Al, including its development, promotion, registration and regulation. The NAIC shall have quasi-judicial powers to hear and decide on cases, and impose administrative sanctions provided under Sec. 20 of this Act. The NAIC shall be responsible for technical support and policy alignment of all government offices, including those under or attached to the DOST and other concerned agencies involved in Al development. The NAIC shall also have the authority to impose administrative penalties in case of any violation of this Act.

Sec. 7. Composition of the NAIC. - The NAIC shall be composed of the

Secretary of the DOST as Chairman, the Secretary of the Department of Information 12 and Communications Technology (DICT) as Vice Chairperson, and the following as members: a) Department of Trade and Industry (DTI); b) Department of Education (DepEd); c) Commission on Higher Education (CHED); d) Department of Labor and Employment (DOLE); e) Technical Education and Skills Development Authority (TESDA); f) Department of Justice (DOJ; g) National Privacy Commission (NPC); h) Department of Economy, Planning, and Development (DepDev); i) National Academy of Science and Technology (NAST); and j) One (1) representative each from the following sectors, to be appointed by the President of the Philippines for a term of three (3) years: i) The private technology sector with expertise in Al or data science; ii) The civil society sector, with expertise in AI ethics, human rights, or digital governance. Each government agency member may designate a permanent representative to the NAIC, with a rank not lower than an Assistant Secretary or its equivalent, and preferably with proven expertise or relevant background in science and technology, information and communications technology, data governance, policy development, or legal and regulatory affairs.

Sec. 8. NAIC Secretariat. - The Secretariat shall implement and execute policies

on Al pursuant to the provisions of this Act. It shall be headed by an Executive Director, with the rank of an Undersecretary. Subject to the review and approval of the Department of Budget and Management (DBM), the NAIC Council shall determine the organizational structure and staffing pattern of the NAIC Secretariat, in accordance with existing Civil Service 7 Commission laws, rules and regulations.

Sec. 9. Powers and Functions of the NAIC. - The functions of the NAIC shall

include, but not limited to, the following: a) Develop and integrate the Philippine AI Roadmap, as provided under Sec. 10 of this Act; b) Coordinate and directly work with all government agencies, local government units, and other stakeholders, whether public and private, involved in AI; c) Maintain a National Registry for AI, as provided under Sec. 11 of this Act; d) Create an AI Ethics Review Board, as provided under Sec. 15 of this Act, which would issue guidelines on Al promotion, sustainable development regulation, safety, ethical standards and accountability; e) Undertake ethical and sustainability reviews of AI systems in coordination with sectoral regulators and ensure that systems align with social, environmental, and intergenerational goals; f) Implement a risk-based regulatory framework of AI systems, including the classification of AI systems as high-risk, moderate-risk, or low-risk based on their potential impact on safety, rights, and national interest; g) Certify and monitor Al-related risks of all AI applications; h) Support Filipino-developed Al technologies through public-private partnerships and targeted innovation grants; i) Coordinate with the National Privacy Commission, Intellectual Property Office, and relevant agencies to ensure Al systems comply with data privacy, intellectual property, and ethical standards; j) Prioritize AI systems that promote the use of AI in agriculture, health, education, disaster risk reduction, governance, environmental sustainability, renewable energy, agriculture, and biodiversity and other sectors;

k) Issue advisory opinions on Al systems that pose unacceptable risks to human rights, public safety, democratic integrity, or environmental sustainability; 1) Create an online portal for the public to access registered AI systems, submit complaints, and report harms; m) Audit, inspect, or suspend Al systems that are not registered or are found to violate the provisions of this Act; n) Hear and decide on complaints, and impose administrative penalties, pursuant to the provisions of this Act; o) Coordinate with international agencies and organizations that have a similar mandate and expertise in artificial intelligence; and P) Perform other functions as may be mandated by law or duly delegated by relevant authorities, as well as those that may be necessary or expedient for the performance of its mandate under this Act.

Sec. 10. Multi-year Philippine Al Roadmap. - The NAIC shall develop and

integrate the multi-year Philippine AI Roadmap that shall guide national and local 16 efforts in harnessing AI. It shall be reviewed at least every three (3) years, or as may 17 be necessary, to adapt to rapidly evolving global Al and technological trends. The NAIC shall ensure the progressive realization of the multi-year Philippine 19 AI Roadmap, which should establish clear targets.

Sec. 11. National Registry of AI Systems. - The NAIC shall establish, maintain,

and regularly update the National Registry of Al Systems. All natural or juridical persons, whether public or private, local or foreign, who develop, deploy, operate, import, sell, or provide access to AI systems in the Philippines, shall be required to register such systems with the NAIC. No AI system shall be imported, sold, deployed, or publicly accessed in the Philippines without prior registration with the NAIC and full compliance with applicable regulatory and ethical standards. The NAIC shall ensure that the registry is maintained in accordance with Republic Act No. 10173, or the Data Privacy Act of 2012, and Republic Act No. 10175, or the Cybercrime Prevention Act of 2012.

Sec. 12. Contents of AI Registration and Registry. - The registry and the

application for AI registration shall include, but not limited to, the following information for each Al system: a) Name and description of the AI system; b) Intended purpose and sector of deployment; c) Name and address of the developer, deployer, or distributor; d) Nature of the AI foundation model (e.g., machine learning, generative Al, LLM); e) Data sources, intended use, and safeguards for training or development, including status or date when the data was last gathered and processed; f) Certification that the Al system has built-in filters or mechanisms to prevent it from being used to facilitate any form of crime, fraud, or harm and that the developers are willing to make the necessary adjustments to the program if the NAIC determines that it is unsafe; g) Risk classification, pursuant to this Act and its implementing rules and regulations; h) Compliance with standards for AI safety, transparency, accountability, explainability, and ethical compliance features, including safeguards against discrimination, misuse, harm, or unintended consequences; i) Mechanism for redress or reporting of adverse incidents arising from its use; and j) Any relevant licenses or certifications issued by national or international authorities. Failure to register or the deliberate concealment of information about an AI system 24 shall be subject to penalties under Sec. 20 of this Act.

Sec. 13. Al Registration as a Prerequisite. - Registration with the NAIC shall be

26 a prerequisite to obtaining any business license or government authorization for the 27 deployment, importation, distribution, or commercial use of an AI system in the country.

Sec. 14. Market Compliance. - All Al systems that were already used, imported,

or accessible in the Philippines prior to the effectivity of this Act shall be registered with the NAIC within sixty (60) days from the effectivity of this Act. Failure to comply shall subject the responsible entity to penalties under this Act.

•

Sec. 15. Al Ethics Review Board. - The NAIC shall create an Al Ethics Review

Board which shall issue implementing guidelines, including detailed risk thresholds, audit mechanisms, and sanctions for non-compliance, aligned with the provisions of this Act. The Al Ethics Review Board shall review risks related to the Al applications and advise on emerging ethical challenges.

Sec. 16. AlI Registration and Risk Classification. - The NAIC shall classify AI

systems and applications according to the following risk tiers: a) High-risk - Al systems' decisions or operations that significantly affect public safety, identity, fundamental rights, livelihoods, or access to essential services, and therefore require heightened regulatory oversight. These include, but not limited to, systems used in education, healthcare, law enforcement, critical infrastructure, justice, finance, and public administration. AI systems classified as high-risk shall undergo mandatory algorithmic impact assessments, data privacy reviews, and sustainability screening before being certified for use. b) Moderate-risk- Al systems that influence outcomes with a limited or reversible effect on individuals, organizations, or social systems, and do not directly compromise safety, rights, or critical services. These include systems used in process automation, service optimization, internal evaluations, and resource allocation. c) Low-risk - AI systems with minimal impact potential, typically used for non- critical tasks such as administrative support, personalized content delivery, or basic data visualization, and which do not materially affect rights, safety, or public interest.

Sec. 17. Responsibilities of AI Developers, Deployers and/or Operators. - All

developers, deployers and/or operators of AI systems shall, include, but not limited to the following: a) Ensure that Al systems are developed and used responsibly, ethically, and transparently; b) Conduct algorithmic audits for bias, security, and privacy; c) Clearly disclose when users are interacting with Al systems;

d) Take full responsibility for harms or unlawful outcomes resulting from their Al system; and e) Provide a clear mechanism to immediately halt, shut down, or disable the Al system if it behaves unpredictably, autonomously exceeds its intended function, causes harm, or commits crime or fraud.

Sec. 18. Al in Employment and Labor Protection. - The NAIC, in coordination

with other concerned agencies, shall regulate the use of AI by employers and business owners, particularly in contexts that may effectively cause worker displacement, job redundancy, or significant changes in labor conditions. It shall likewise issue a separate guidelines, and shall have, but not limited to, the following powers and responsibilities: a) Al Job Classification Standards - integrate Al-related roles into occupational classification frameworks, which may include, but not limited to, the following: Define Al-related occupations, roles, and skill levels across sectors (e.g., healthcare, finance, government, business process outsourcing, manufacturing). Classify roles based on, skill level, education and training requirements, and certifications (e.g., from TESDA, CHED, private institutions); and iii) Link AI Job Classifications to Wage, Labor Standards, and Employment Services. b) Advance Notification and Impact Assessment - require all employers intending to adopt or integrate AI systems, which may result in displacement, retrenchment, or redundancy of employees to: i) Submit a prior AI Deployment Impact Report detailing the nature of the AI system, affected job functions, and estimated number of workers at risk; and ii) Notify the DOLE at least sixty (60) days prior to implementation. c) Regulation and Conditional Use - develop a regulatory framework that will mitigate the possible adverse effect to employees of the use of Al systems in workplaces, including but not limited to: i) Mandatory retraining, upskilling, or redeployment programs; ii) Transition assistance packages for affected employees; and

ili) Proof of employer engagement with displaced workers on alternative employment pathways. d) Tripartite Oversight and Consultation - institutionalize a Tripartite Al and Labor Transition Council, composed of representatives from the DOLE, employers' organizations, labor groups, and civil society, to: i) Review Al-related employment trends; ii) Recommend policy safeguards to DOLE and NAIC; and iii) Facilitate labor-management dialogue in sectors undergoing Al transformation. e) Skills Mapping and Future-Proofing - conduct regular labor market evaluation to identify job categories at risk of automation and promote the following: 1) Proactive skills development aligned with emerging Al demands; ii) Support for technical-vocational education, apprenticeships, and digital skills training in collaboration with TESDA, CHED, and DepEd; and iii) Government-led or subsidized upskilling for workers, especially in high- risk industries. f) Employment Guarantee Mechanisms - develop and pilot employment guarantee schemes or public sector work programs for displaced workers due to Al-induced changes in their employment. g) Monitoring and Enforcement - develop a system for reporting and auditing Al- induced layoffs, with sanctions for non-compliance including administrative fines, revocation of business permits, or disqualification from government incentives.

Sec. 19. Al and Sustainable Innovation Hubs. - Regional AI and Sustainable

Innovation Hubs shall be established in partnership with the NAIC, DOST research centers, state universities and colleges (SUCs), and NAIC-accredited private institutions. The NAIC shall also support the creation of specialized training tracks, research laboratories, public-private innovation hubs, and startup incubation programs to encourage the growth of homegrown AI talent and Filipino-developed AI technologies. Capacity-building efforts shall be inclusive, accessible, and aligned with national

development priorities. These hubs shall also promote innovation in the sectors of government.

Sec. 20. Prohibited Acts and Penalties. - Any person, whether natural or

juridical, who commits any of the prohibited acts, without prejudice to civil and other criminal liabilities under existing laws, shall be subject to the following penalties: a) Development or Deployment of Unregistered Al Systems - any person who develops, deploys, distributes, or makes commercially available any AI system without prior registration with the NAIC shall be punished with: i) Revocation of business permit or licenses, and blacklisting from government procurement and Al development grants; and Impose a fine of Five Hundred Thousand Pesos (Php 500,000) to Five Million Pesos (Php 5,000,000) or imprisonment of six (6) months to three (3) years, or both, at the discretion of the court; b) Use of Al Systems to Commit, Facilitate, or Conceal Fraud, Crimes, or Cause Harm - Any person who intentionally uses Al systems to commit fraud, facilitate or conceal crimes, or cause harm to life, liberty, property, or national security shall suffer: i) Impose a fine of Two Million Pesos (Php 2,000,000) to Ten Million Pesos (Php 10,000,000) or imprisonment of six (6) years to twelve (12) years, or both, at the discretion of the court; and ii) If the offense results in death, physical injury, or large-scale financial or reputational damage, penalties shall be imposed in their maximum period. c) Violation of Safeguards under this Act, or Rules and Regulations Issued by the NAIC or the Al Ethics Review Board - Any person or entity that willfully violates administrative rules, technical standards, ethical guidelines, or reporting requirements issued by the NAIC or the Al Ethics Review Board shall be liable for: i) An administrative fine of One Hundred Thousand Pesos (Php 100,000) to One Million Pesos (Php 1,000,000) per violation; ii) Suspension or revocation of registration or certification of the Al system involved; and

iii) Mandatory compliance training or ethics audit as a condition for reinstatement or continued operation. d) Failure to Disclose Al-Generated Content or Provide Required Disclaimers - Any person who knowingly fails to label or disclose Al-generated content, particularly in sensitive contexts (e.g., political, medical, educational, or legal) shall be penalized with: i) Impose a fine of Three Hundred Thousand Pesos (Php 300,000) to One Million Pesos (Php 1,000,000) or imprisonment of six (6) months to two (2) years, or both, at the discretion of the court; and Platforms or publishers may be held administratively liable for failure to implement monitoring or compliance protocols; e) Negligent Operation of Harmful or Malfunctioning AI Systems - Any developer or deployer who knowingly allows an Al system to continue operating after it has exhibited harmful, unlawful, or unpredictable behavior shall be punished with: i) Impose a fine of Five Hundred Thousand Pesos (Php 500,000) to Three Million Pesos (Php 3,000,000) or imprisonment of one (1) year to five (5) years, or both, at the discretion of the court; and = Permanent disqualification from holding any license to develop or operate Al systems in the Philippines, subject to due process. f) Use of Al to Manipulate Public Opinion, Spread Disinformation, or Conduct Unlawful Surveillance - Any person who uses Al to create or disseminate disinformation, conduct mass opinion manipulation (e.g., through bots, deepfakes), or carry out surveillance without legal authority shall be punished with: Impose a fine of One Million Pesos (Php 1,000,000) to Five Million Pesos (Php 5,000,000), or imprisonment of three (3) years to ten (10) years, or both, at the discretion of the court; and Additional penalties shall apply if such acts are committed during election periods, public emergencies, or in violation of constitutional rights; g) Misrepresentation of AI-Generated Content as Human-Made - Knowingly passing off Al-generated content (e.g., fake images, news articles,

endorsements, voice recordings) as human-made for purposes of deceit, manipulation, or commercial gain shall be penalized with: i) Impose a fine of Five Hundred Thousand Pesos (Php 500,000) to Two Million Pesos (Php 2,000,000) or imprisonment of six (6) months to three (3) years or both, at the discretion of the court. The penalties above shall be imposed in their maximum period when any of the following circumstances are present: a) The offender is a public officer or government employee abusing official position; b) The violation targets minors, senior citizens, or other vulnerable groups; c) The offense is committed in relation to an electoral exercise or during a state of national emergency; or d) The Al system was used to commit multiple or repeated offenses.

Sec. 21. Appropriations. - The amount necessary for implementation of this Act

shall be charged against the current year's appropriations of the DOST. Thereafter, such sums as may be necessary for the continued implementation shall be included in the annual General Appropriations Act.

Sec. 22. Implementing Rules and Regulations. - Within ninety (90) days from

the effectivity of this Act, the DOST, DICT, and other relevant agencies shall promulgate the implementing rules and regulations.

Sec. 23. Separability Clause. - If any portion or provision of this Act is declared

unconstitutional, the remainder of this Act or any provisions not affected thereby shall remain in force and effect.

Sec. 24. Repealing Clause. - All law, presidential decree or issuance, executive

order, letter of instruction, rule of regulation inconsistent with the provisions of this 26 Act is hereby repealed or modified accordingly.

Sec. 25. Effectivity. - This Act shall take effect after fifteen (15) days following

its complete publication in the Official Gazette or the newspaper of general circulation. Approved,