On April 13, 2026, the Philippines' Department of Justice (DOJ), Presidential Communications Office (PCO), and Department of Information and Communications Technology (DICT) signed a memorandum of agreement launching Oplan Kontra Fake News, a whole-of-government framework against online disinformation and AI-generated deepfakes. The division of labor is clean: the PCO runs media-literacy and counter-disinformation campaigns, the DICT handles digital reporting systems and cybersecurity support, and the DOJ evaluates cases for prosecution under existing laws, with an inter-agency steering committee coordinating evidence handling. "In the age of artificial intelligence and hyper-speed information, truth must move faster than deception," DICT Secretary Henry Aguda said at the signing, per Newsbytes.ph.
Two of those three lanes are exactly what proportionate policy looks like. The third — prosecution of an offense no Philippine statute squarely defines — is where the framework will be tested.
The threat is real, and the steelman is strong
Manila's alarm is earned, not manufactured. Sumsub's Asia-Pacific fraud research, reported by Fintech News Singapore, found the Philippines posted the region's highest year-on-year increase in synthetic identity-document fraud in the first quarter of 2025 — 291%, against an APAC-wide rise of 233%. The country has already run one national election under deepfake conditions: ahead of the May 2025 midterms, the Commission on Elections (COMELEC) had to build an entire task force to police AI-manipulated campaign content. With another election cycle approaching in 2026, a government that simply waited for synthetic media to corrode public trust would be negligent. Coordinating three agencies that each own a genuine piece of the problem — communication, technology, law — is good administration, not censorship.
The PCO lane builds on real groundwork: the Philippine News Agency reports that on March 3, 2026, the PCO signed a memorandum of understanding with the publishers and editors of the nine national broadsheets to support the campaign. Media literacy and faster official communication are speech-expanding tools. The DICT's forensics and reporting infrastructure likewise targets capability, not content. Neither lane requires anyone to decide what citizens may say.
The DOJ lane prosecutes an undefined noun
The problem is the prosecutorial third. The Philippines has no statute that defines "fake news" as a general offense. What prosecutors actually have are adjacent instruments: Article 154 of the Revised Penal Code, an 86-year-old provision punishing publication of false news that endangers public order, and cyber libel under the Cybercrime Prevention Act of 2012 (Republic Act No. 10175), which the Supreme Court upheld in Disini v. Secretary of Justice (2014) with penalties one degree higher than offline libel. That toolkit has a documented history of being aimed at journalists — most famously the 2020 cyber-libel conviction of Nobel laureate Maria Ressa.
When a framework's enforcement arm runs on elastic statutes and an undefined target, the operative question becomes who gets charged first. Newsbytes.ph noted at the signing that civil-society observers see no operational clarity yet on how the steering committee will distinguish disinformation from protected speech, satire, or simple error. The Constitution's free-speech clause — Article III, Section 4 — is not a technicality here; it is the reason "fake news" prosecutions without precise definitions tend to chill far more lawful speech than they deter lies.
Manila already wrote a better model — so did Brussels
The irony is that the Philippines has produced one of the developing world's better templates for handling synthetic media. COMELEC Resolution No. 11064, promulgated September 17, 2024, regulated AI in the 2025 campaign with everything Oplan Kontra Fake News currently lacks: precise definitions (it defines "deepfakes" and even "cheapfakes"), a bounded context (digital election campaigning only), a designated implementing arm (Task Force KKK sa Halalan), and anchoring in a specific pre-existing offense — Section 261(z)(11) of the Omnibus Election Code. It even recites the constitutional free-speech guarantee in its preamble. It regulates the artifact — undisclosed synthetic campaign media — not the truthfulness of opinions.
The European Union has just taken the same artifact-first path at scale. On June 10, 2026, the European Commission published its Code of Practice on the marking and labelling of AI-generated content under Article 50 of the AI Act, ahead of transparency obligations that bite on August 2, 2026. As MediaNama's analysis details, deployers must label deepfakes clearly and at first exposure, using standardized, empirically tested icons. The EU's answer to "lies can look real" is disclosure and machine-readable provenance — not asking prosecutors to adjudicate truth.
Define it before you charge it
The constructive path for Manila is clear. Congress should give the DOJ the narrow instrument it currently lacks: a deepfake statute built on intent and harm — impersonation, fraud, electoral deception — with explicit safe harbors for satire, parody, and journalism, mirroring the definitional discipline COMELEC already showed. Until then, the Oplan Kontra Fake News steering committee should publish its prosecution guidelines: which statutes it will invoke, what evidentiary thresholds apply, and a public log of referrals and outcomes. An anti-disinformation body that is transparent about its own operations earns the credibility it needs.
Two-thirds of this framework — literacy and forensics — is the proportionate playbook other emerging democracies should copy. The remaining third is running on borrowed statutes and an undefined noun. The lesson from both Intramuros and Brussels is the same: regulate the synthetic artifact, define the offense precisely, and keep prosecutors out of the truth business.