Framework for Accountable and Intelligent Regulation of Artificial Intelligence (Fair) Act

SenaTe Office of the Secretary TWENTIETH CONGRESS OF THE ) REPUBLIC OF THE PHILIPPINES First Regular Session MAY -5 P3:20 SENATE S. No. 20977 RECEIVED BY: Introduced by Senator Camille A. Villar AN ACT INSTITUTIONALIZNG A COMPREHENSIVE REGULATORY FRAMEWORK ON THE USE OF ARTIFICIAL INTELLIGENCE EXPLANATORY NOTE Under the 1987 Constitution, society is envisioned wherein technology serves human dignity, intellectual creation is protected, and economic activity advances the common good. However, the rise of Artificial Intelligence (AI) threatens these principles. Deepfakes and AI-generated disinformation tarnishes reputations, manipulate public opinion, and erode trust in the media. Al systems trained on copyrighted works without authorization deprive Filipino creators of their intellectual property rights. The growing preference for mass-produced AI-generated media displaces and undervalues Filipino artists. This bill establishes a comprehensive regulatory framework that balances innovation with respect to humanity. Malicious synthetic media intended to deceive is met with penal sanctions. Commercial Al usage is permitted but conditioned upon strict disclosure and accountability mechanisms. Digital platforms bear clear obligations to address violative content, and users accountable for Al-generated outputs. It positions the Philippines as a responsibly regulated player in the global digital economy, ensuring AI serves human progress and development. In view of the forgoing, the approval of this measure is urgently sought. CAMILLE A. VILLAR

senate Office of the Secretare TWENTIETH CONGRESS OF THE REPUBLIC OF THE PHILIPPINES ) 26 MAY -5 P 3:20 First Regular Session S. No. SON 4057 RECEIVED BY: Introduced by Senator Camille A. Villar AN ACT INSTITUTIONALIZING A COMPREHENSIVE REGULATORY FRAMEWORK ON THE USE OF ARTIFICIAL INTELLIGENCE Be it enacted by the Senate and the House of Representatives of the Philippines in Congress assembled: CHAPTER I GENERAL PROVISIONS

Section 1. Short Title. — This Act shall be known as the "Framework for

2 Accountable and Intelligent Regulation of Artificial Intelligence (FAIR-AI) Act".

Sec. 2. Declaration of Policy. - It is hereby declared the policy of the State to

establish a risk-proportionate regulatory framework governing the development, 5 deployment, and use of Artificial Intelligence (AI) across all public and private sectors. 6 AI systems shall operate transparently, respect constitutional rights, maintain verifiable 7 audit trails, and be subject to accountability measures that prevent deception, 8 discrimination, privacy violations, and threats to public safety. Non-commercial, 9 personal, academic, and purely artistic uses of AI are expressly excluded from criminal and regulatory liability under this Act, except when such uses cause material harm to third parties, involve public sector deployment, or are integrated into critical infrastructure or essential services.

Sec. 3. Scope of Application. — This Act shall apply to all natural and juridical

14 persons, including government agencies, local government units, state universities and colleges, and government-owned or controlled corporations, engaged in the development, training, fine-tuning, deployment, or use of AI systems in:

a) Public Sector - Including but not limited to government service delivery, policy formulation, public administration, law enforcement, national security, and judicial support systems; b) Essential services - Including but not limited to healthcare delivery, public education, social welfare, water and energy utilities, and emergency response; c) Critical infrastructure - Including but not limited to transportation systems, telecommunications, financial infrastructure, and industrial control systems; d) Commerce, Trade, and Industry - Including but not limited to advertising, manufacturing, logistics, financial services, e-commerce, and employment systems; e) Media and Communications - Including but not limited to news production, content distribution, social media platforms, and public information campaigns; f) Research and Development - Including but not limited to scientific research, public health surveillance, and academic AI applications with public impact. This Act shall not apply to AI used exclusively for personal, non-commercial, educational, or independent artistic creation unless such use: (1) involves public sector deployment; (2) interfaces with critical infrastructure; (3) processes personal data at scale; or (4) causes material, demonstrable harm to identifiable persons or public interests.

Sec. 4. Definition of Terms. — For purposes of this Act, the following definitions

shall apply: a) Artificial intelligence (AI) - refers to software systems or computational processes capable of performing tasks requiring pattern recognition, predictive modeling, natural language processing, or content generation through machine learning, neural networks, or algorithmic inference, without relying solely on explicit, step-by-step human programming for each output. b) Al-generated content - refers to any text, audio, image, video, code, or data output produced or substantially modified by an AI system, regardless of the degree of human prompting, editing, or oversight.

c) Automated Decision System — refers to an AI system that makes or substantially influences binding determinations affecting legal rights, access to services, employment, benefits, or liberty, with or without subsequent human review. d) Text-and-Data Mining (TDM) refers to the automated computational analysis of large-scale copyrighted or publicly available works to extract patterns, statistical relationships, or structural insights for AI model training, subject to applicable intellectual property and fair use provisions. e) Original Work — refers to an intellectual creation protected under Republic Act No. 8293, otherwise known as "Intellectual Property Code", reflecting substantial human creative direction, selection, or arrangement. Purely AI-generated outputs without meaningful human creative input shall not qualify as original works under this definition, consistent with Section 2 of Republic Act No. 8293, which requires authorship by a natural person. f) Low Risk Al Use - refers to AI deployment with minimal direct impact on individual rights, public safety, or market integrity, including but not limited to internal business automation, non-deceptive creative or artistic experimentation, educational or illustrative content clearly labeled as such, and personal productivity tools. g) Medium-Risk AI Use - refers to AI deployment with direct public-facing or commercial impact that does not determine legal rights, essential services, or critical infrastructure, including but not limited to marketing and advertising, customer service automation, content distribution for profit, and news commentary or opinion generation. h) High Risk Al Use — refers to AI deployment in contexts where errors, bias, or manipulation could materially affect constitutional rights, public safety, financial stability, health outcomes, or critical operations, including but not limited to automated eligibility determinations, financial credit or insurance underwriting, medical diagnosis or treatment planning, public safety or emergency response systems, and critical infrastructure control. i) Al Value Chain - refers to the sequential or interconnected roles involved in the AI lifecycle, comprising foundation model developers, fine-tuners, application

developers, deployers, distributors, digital platforms, and end-users, for purposes of allocating accountability under this Act. j) Foundational Model Developer - refers to an entity that designs, trains, and distributes large-scale models intended for broad or downstream commercial, public, or industrial use. k) Fine-Tuner — refers to an entity that adapts, retrains, or modifies a foundation model or its parameters, prompts, or architectures for specific tasks, domains, or operational contexts. I) Application Developer — refers to an entity that builds software, interfaces, or integrated workflows that utilize Al models to deliver end-user or enterprise functions. m) Distributor - refers to any entity that commercially licenses, markets, or resells AI models, tools, or AI-generated content to third parties. n) Deployer - refers to any natural or juridical person, including government agencies, local government units, state-owned or controlled corporations, and private enterprises, that operationalizes or integrates AI systems into commercial, industrial, public service, or administrative workflows. o) Digital Platform - refers to an online service, including social networks, content aggregators, streaming services, and e-commerce marketplaces, that hosts, distributes, or monetizes user-uploaded or Al-generated content and exercises editorial, algorithmic, or technical control over its presentation, accessibility, or distribution. p) End-User - refers to the final recipient or consumer of Al-assisted products, services, or content. End-users shall not be held liable under this Act unless they modify, redistribute, or deploy AI-generated content for commercial, public, or high-impact purposes. q) Machine-Readable Disclosure Metadata - refers to standardized, structured data tags (e.g., JSON-LD, C2PA-compliant schemas, or DICT-published formats) embedded in Al outputs to verifiably indicate Al involvement, risk classification, deploying entity, and redress contact.

r) Training Data Provenance Log — refers to a machine-verifiable, tamper-evident record documenting the source, licensing status, opt-out compliance, preprocessing steps, and retention period of datasets used to train or fine-tune an Al system. 5) Opt-Out Mechanism - refers to a technically feasible, machine-readable, and publicly accessible method allowing rights-holders to prevent their copyrighted works or identifiable likeness from being used in AI training or generation. t) Malicious Synthetic Media - refers to AI-generated or Al-manipulated audio, visual, or textual content that intentionally simulates a real person without consent, or deceptively presents fabricated events as factual, with the purpose of causing reputational, financial, physical harm, or fraud. u) Regulatory Sandbox - refers to a time-bound, supervised testing environment authorized by the Council, created under Section 6 of this Act, wherein eligible entities may pilot novel AI applications under modified compliance requirements, subject to predefined risk mitigation, monitoring, and exit protocols. CHAPTER II OPERATIONAL PRINCIPLES AND TECHNICAL STANDARDS

Sec. 5. Operational Principles. - In implementing this Act, all regulatory actions,

technical standards, and enforcement decisions shall be guided by the following binding principles: a) Human Oversight - Automated Decision Systems affecting legal rights, essential services, or public safety shall maintain verifiable human-review audit trails and appeal mechanisms; b) Verifiable Transparency - Al involvement in public-facing or commercial content shall be disclosed through machine-readable, technically interoperable metadata conforming to DICT standards; c) Risk-Proportionate Accountability - Compliance obligations shall scale with the potential impact of Al deployment, with enhanced safeguards for High-Risk AI Use as defined under Section 4(h) of this Act;

d) Fairness and Non-Discrimination — High-Risk AI Uses shall undergo bias impact assessment prior to deployment; deployers shall implement reasonable mitigation measures where systemic discrimination affecting protected classes is identified. e) Safety, Security, and Resilience - AI systems controlling or interfacing with critical infrastructure, emergency response, or public safety operations shall incorporate human-override capability, incident response protocols, and annual resilience reporting to DICT. f) Innovation and Economic Enablement - The State shall foster responsible innovation by maintaining regulatory sandboxes, issuing advisory guidelines for micro, small, and medium enterprises, and ensuring that compliance costs do not disproportionately burden emerging Filipino developers and creators. g) Adaptive Governance - Technical standards, disclosure schemas, and risk- classification methodologies shall be reviewed by DICT in consultation with relevant stakeholders; updates shall be published through the IRR process without requiring legislative amendment, provided they remain within the scope and intent of this Act. h) Legal Compliance - All AI systems shall operate in conformity with Republic Act No. 8293 (Intellectual Property Code of the Philippines), Republic Act No. 10173 (Data Privacy Act of 2012), and other applicable laws. CHAPTER III PHILIPPINE AI REGULATION COUNCIL (PARC)

Sec. 6. Creation, Composition, and Membership. — The Philippine AI Regulation

Council (PARC), hereinafter referred to as the Council, is hereby established. The composition of the Council shall be as follows: a) Ex-Officio Members — The following executive departments and agencies shall serve as voting members by virtue of their office: i. Department of Information and Communications Technology (DICT) — Chairperson

ii. Department of Economy, Planning, and Development (DEPDev) - Member ili. Department of Trade and Industry (DTI) - Member iv. Department of Finance (DOF) - Member V. Department of Interior and Local Government (DILG) - Member vi. Department of Justice (DOJ) — Member vii. Intellectual Property Office of the Philippines (IPOPHL) - Member vili. National Privacy Commission (NPC) - Member b) Advisory Members - The following executive departments and agencies shall designate representatives to serve in an advisory capacity, with the right to participate in deliberations, submit technical recommendations, and attend working committees: i. Bangko Sentral ng Pilipinas (BSP) ii. Bureau of Philippine Standards (BPS) iii. Civil Service Commission (CSC) iv. National Youth Commission (NYC) V. Department of Education (DepEd) vi. Department of Health (DOH) VII. Department of Energy (DOE) viii. Department of Public Works and Highways (DPWH) c) Permanent Consultants - The Council may, by resolution, appoint technical experts from academia, industry associations, developer networks, or civil society as permanent consultants to provide specialized input on standards development, risk assessment methodologies, or sector-specific guidelines.

Sec. 7. Functions. - The Council shall exercise the following functions:

a) Develop and publish technical standards for Machine-Readable Disclosure Metadata, Training Data Provenance Logs, disclosure protocols, labeling specifications, and commercially feasible watermarking mechanisms, in consultation with industry, technical experts, the Bureau of Philippine Standards, and civil society;

b) Maintain and update a national registry of High-Risk Al Systems, Automated Decision Systems, and regulatory sandbox participants; c) Conduct technical audits, compliance reviews, and risk assessments of Al deployments across all sectors, and issue compliance directives or corrective measures as necessary; d) Operate a centralized, audit-tracked grievance portal and incident reporting system, and coordinate with relevant government bodies for investigation, resolution, or criminal referral; e) Develop, publish, and periodically review Section f) Establish, administer, and monitor a Regulatory Sandbox program to enable time- bound, supervised testing of novel Al applications under modified technical compliance requirements, while preserving liability for intentional harm, fraud, or criminal violations; g) Assess the macroeconomic, fiscal, employment, and developmental impacts of AI regulation, ensuring that technical standards, disclosure obligations, and compliance requirements are proportionate, budgetarily feasible, and aligned with national competitiveness goals; h) Coordinate with relevant government agencies to ensure integrated oversight of data privacy, criminal enforcement, intellectual property, consumer protection, public safety, financial stability, education, and public sector administration; Issue technical guidelines defining risk classifications, material harm thresholds, commercially feasible mitigation measures, and sector-specific compliance methodologies, subject to biennial review and public publication; j) Publish anonymized incident trends, policy recommendations, and regulatory impact assessments to inform continuous improvement of Al governance frameworks; and k) Perform such other functions as may be necessary to effectively implement the provisions of this Act.

Sec. 8. Decision-Making and Quorum. - A majority of the voting members shall

constitute a quorum. Decisions shall be made by majority vote of voting members

1 present. In case of a tie, the Chairperson shall cast the deciding vote. Advisory members 2 and permanent consultants may participate in deliberations.

Sec. 9. Mandatory Technical Compliance Code (MTCC) for AI Value Chain. — The

4 Council shall develop and promulgate a Mandatory Technical Compliance Code (MTCC) 5 within one hundred twenty (120) days from the effectivity of this Act. Such Code shall 6 have binding legal effect and shall apply exclusively to: (a) Foundation Model 7 Developers; (b) Fine-Tuners; (c) Application Developers; (d) Deployers; (e) Distributors; and (f) Digital Platforms, as defined under Section 4, paragraphs (i), (k), (I), (m), (n) and (o) of this Act. The Code shall not apply to End-Users as defined in Section 4(p), 10 academic researchers, or entities engaged in non-commercial, personal, educational, or purely artistic Al uses. The MTCC shall include: a) Binding technical and operational standards for disclosure, metadata embedding, labeling, and human oversight; b) Required risk assessment and mitigation protocols for High-Risk AI Use; c) Mandatory incident detection, containment, and reporting procedures; d) Compliance frameworks for honoring registered opt-out mechanisms for copyrighted works and personal data; and e) Value-chain documentation, audit, and inter-actor cooperation obligations. Compliance with the MTCC shall be required for all covered actors. Documented adherence to the Code shall be considered a mitigating circumstance in determining administrative penalty, but shall not constitute a defense against liability for intentional violations, provided that: (a) the actor maintained verifiable records of compliance efforts; (b) the violation did not involve intentional, fraudulent, or malicious conduct; and (c) the deficiency was promptly remedied upon notice. Failure to comply with the Code shall constitute an administrative violation subject to notices, directives for compliance, and suspension orders. The Code shall be published in the Official Gazette and on the PARC website within thirty (30) days of promulgation.

Sec. 10. Grievance and Reporting Mechanism. — The Council shall establish and

maintain a centralized, accessible, and secure Artificial Intelligence (AI) Grievance and

1 Incident Reporting Portal to facilitate the reporting of violations, harms, or risks associated with AI systems. The mechanism shall include the following: a) Accessible Reporting Channels. — The Portal shall be accessible via web interface, mobile application, and offline channels to ensure inclusivity. Reports may be filed by any natural or juridical person. b) Categories of Reportable Incidents. — The Portal shall allow users to categorize reports into, but not limited to: 1. Undisclosed AI-generated content; 2. Intellectual property infringement or unauthorized use of protected works; 3. Malicious synthetic media, deepfakes, or non-consensual likeness replication; 4. Bias, discrimination, or unfair outcomes from Automated Decision Systems; 5. Data privacy violations involving Al processing; and 6. Safety failures in High-Risk AI systems. c) Assessment Protocol. — Upon receipt of a report, the Council, through an interdisciplinary technical team, shall: 1. Acknowledge receipt within forty-eight (48) hours; 2. Conduct an assessment to determine the severity and risk level of the incident; and 3. For urgent cases involving imminent harm to public safety, national security, or severe reputational damage, initiate an expedited review within twenty-four (24) hours. d) Coordination with Appropriate Agencies. - The Council shall coordinate with appropriate agencies based on the nature of the complaint. e) Remedial Actions and Redress. - Depending on the findings, the Council may issue the following: 1. Direct digital platforms or deployers to remove or disable access to violative content within a prescribed period; 2. Require the publication of corrections or clarifications where misinformation has caused public harm;

3. Temporarily suspend the operation of an AI system pending further investigation if it poses an immediate threat to public safety or rights; and 4. Forward evidence of criminal liability to the DOJ for appropriate legal action. f) Whistleblower Protection. - Individuals who report violations in good faith, particularly employees or insiders of Al developers, deployers, or digital platforms, shall be protected from retaliation, dismissal, or harassment under existing whistleblower protection laws. The identity of reporters shall remain confidential unless disclosure is required by court order. g) Transparency and Feedback. — The Council shall publish anonymized statistics on the number of reports received, types of violations, actions taken, and resolution timelines on a quarterly basis. Reporters shall be informed of the status and outcome of their complaints, subject to confidentiality and legal constraints.

Sec. 11. Congressional Oversight and Reporting. — To ensure accountability and

facilitate legislative review of the implementation of this Act, the Council shall submit reports to the Congress as follows: a) Annual Comprehensive Report. — Within sixty (60) days after the end of each calendar year, the Council shall submit an annual report to the Senate and the House of Representatives. The report shall include, but not be limited to: 1. The number of AI systems registered in the national registry, categorized by risk level (Low, Medium, High); 2. A summary of complaints received through the grievance portal, investigations conducted, compliance directives issued, and penalties imposed; 3. An analysis of emerging AI-related risks; 4. An assessment of the impact of AI regulation; 5. A statement of expenditures and utilization of funds appropriated for the Council; and 6. Proposed amendments to this Act or related laws based on technological advancements and regulatory gaps identified during the year.

b) Special Reports. — The Council shall immediately submit special reports to the appropriate committees of both the Senate and House of Representatives upon the occurrence of any of the following: 1. A widespread or systemic failure of High-Risk AI systems affecting public safety, national security, or critical infrastructure; 2. A significant breach of data privacy or intellectual property rights involving AI systems that affects a large number of Filipinos; 3. Upon request by any committee within thirty (30) days from receipt of such request. c) Public Availability. - All reports submitted to Congress, excluding confidential information protected by law or involving ongoing investigations, shall be made publicly available on the Council's website within fifteen (15) days from submission to ensure transparency and public awareness. d) Congressional Hearings. - The Chairperson of the Council, or their duly authorized representative, shall appear before the relevant committees of both the Senate and House of Representatives at least once a year, or as often as required, to discuss the annual report, address legislative concerns, and provide updates on the state of Al governance in the Philippines. CHAPTER IV PROHIBITED ACTS AND PENALTIES

Sec. 12. Prohibited Acts. - It shall be unlawful for any natural or juridical person

to commit the following acts: a) Undisclosed Distribution of Medium-Risk or High-Risk AI Content - The sale, lease, licensing, or public distribution of Al-generated content classified as Medium-Risk or High-Risk under Section 4(g)-(h) without affixing the clear, conspicuous, and machine-readable disclosure mandated under Section 7 and the Mandatory Technical Compliance Code. Low-Risk AI uses under Section 4(f) are exempt from pre-distribution disclosure but remain subject to post- deployment notification. b) Infringing Use of Source Material - The utilization of Al systems to train, fine- tune, or generate content that substantially reproduces protected elements of an

Original Work as defined under Republic Act No. 8293 (Intellectual Property Code), without authorization from the rights-holder, where the actor: (1) acted with actual knowledge of the infringement; or (2) failed to maintain a documented Training Data Provenance Log or honor registered opt-out mechanisms under Section 4(r). c) Misrepresentation of Origin - The false representation that AI-generated or AI- assisted content was produced exclusively by human creators. d) Circumvention of Technical Safeguards - The intentional removal, alteration, encryption, masking, obfuscation, or technological circumvention of any AI disclosure label, watermark, metadata schema, or machine-readable identifier applied to content pursuant to this Act, its Implementing Rules and Regulations (IRR), or the Mandatory Technical Compliance Code (MTCC). e) Creation and Distribution of Malicious Synthetic Media - The generation, publication, or dissemination of AI-manipulated audio, visual, or textual content that: (1) intentionally simulates a person without their prior informed, explicit consent; or (2) deceptively presents fabricated events as factual; with the purpose of causing reputational, financial, or physical harm, defrauding any person or entity, or undermining public safety, consistent with the prohibitions under Republic Act No. 10175 or Cybercrime Prevention Act of 2012. f) Unauthorized Likeness or Voice Replication - The use of AI systems to synthesize, clone, or simulate the identifiable voice, facial likeness, or biometric markers of a natural person for exploitation, harassment, or deception without prior informed, explicit consent. g) Digital Platform Non-Compliance - The failure by a digital platform to maintain accessible reporting mechanisms, act upon verified notices from PARC within prescribed timelines, publish required transparency reports, or implement reasonable content moderation protocols as mandated under Section 8. h) Circumvention of Risk Classification - The misclassification of a high-risk Al use as low or medium-risk.

Sec. 13. Liability of Juridical Entities and Responsible Officers. - Any natural

person who is a director, officer, manager, or other individual exercising supervisory or

1 decision-making authority over the Al system or activity in violation of this Act shall be held jointly and severally liable with the juridical entity only for fines and restitution, but 3 shall additionally bear individual criminal liability separate from the entity. A responsible 4 officer is liable under this section if: a) The officer authorized, directed, consented to, or knowingly failed to prevent the violation; b) The officer acted with gross negligence, recklessness, or willful disregard of the requirements of this Act; or c) The officer had actual knowledge of the violation and failed to take reasonable corrective action within a prudent timeframe after such knowledge. The criminal prosecution of a juridical entity shall not bar the separate prosecution of any responsible officer, and vice versa. A finding of criminal liability against the officer may be made even if the juridical entity is not prosecuted or cannot be penalized for reasons such as dissolution, insolvency, or other legal disability. The penalties imposable upon a responsible officer shall be the same as those prescribed under Section 14 of this Act for a natural person, without reduction, and shall be served personally. The dissolution, liquidation, merger, or reorganization of a juridical entity after the commission of a violation shall not extinguish its liability for fines and sanctions under this Act. The successor-in-interest shall be liable for payment of fines and compliance with remedial orders.

Sec. 14. Fines and Penalties. - For any act prohibited under Section 12, the

following fines and penalties shall be imposed upon conviction: a) Violation of Section 12(a), (d), (g), (i), or (i) shall be punished by a fine of not less than Two Hundred Fifty Thousand Pesos (Php250,000.00) but not more than Seven Hundred Fifty Thousand Pesos (Php750,000.00), or imprisonment of not less than six (6) months but not more than two (2) years, or both. b) Violation of Section 12(b), (c), or (h) shall be punished by a fine of not less than Seven Hundred Fifty Thousand Pesos (Php750,000.00) but not more than Two Million Pesos (Php2,000,000.00), or imprisonment of not less than two (2) years but not more than four (4) years, or both.

c) Violation of Section 12(e) or (f) shall be punished by a fine of not less than Two Million Pesos (Php2,000,000.00), or imprisonment of not less than four (4) years but not more than eight (8) years, or both. CHAPTER V FINAL PROVISIONS

Sec. 15. Appropriations. - The amount necessary for the initial implementation

of this Act shall be taken from the current appropriations of each of the member- agencies of the Council. Thereafter, the amount needed for the continued 10 implementation of this Act shall be included in the annual appropriations of the Council 11 as a distinct and separate item.

Sec. 16. Implementing Rules and Regulations. - Within ninety (90) days from

the effectivity of this Act, the Council shall promulgate its implementing rules and regulations (IRR). All guidelines, regulations, and other issuances mandated to be promulgated under this Act shall be issued within sixty (60) days from the promulgation of the IRR.

Sec. 17. Separability Clause. - If any provision of this Act is declared invalid or

unconstitutional, the remaining provisions shall continue to be in full force and effect.

Sec. 18. Repealing Clause. - All laws, decrees, executive orders, proclamations,

rules and regulations, and issuances, or parts thereof which are inconsistent with the provisions of this Act, are hereby repealed, amended, or modified accordingly.

Sec. 19. Effectivity. - This Act shall take effect after fifteen (15) days following

its publication in the Official Gazette or in two (2) newspapers of general circulation. Approved,