What the Philippine Supreme Court's AI framework means for your firm

Most courts writing about AI have written warnings. In February 2026 the Philippine Supreme Court wrote something closer to a manual. Its Governance Framework on the Use of Human-Centered Augmented Intelligence in the Judiciary reaches court users, including members of the Bar, and it is unusually practical.

It asks for three things. Disclose the AI tools you used, in plain language, along with the degree of AI involvement and the human oversight applied. Make that use auditable and traceable. And accept that responsibility stays with you; you cannot escape it by blaming the tool. Notably, it lists redaction of data from documents among the AI uses it endorses.

So the highest court in the country named a specific task, redaction, as a legitimate use of AI, and then asked for disclosure, oversight, and a traceable record around it. That is a workflow, and it maps almost exactly onto what a careful firm would want anyway.

Underneath the framework sits the Data Privacy Act. Republic Act No. 10173 treats government identifiers as sensitive personal information, and the National Privacy Commission enforces it actively, with penalties that reach seven years and several million pesos depending on the offense. Recent issuances apply the Act across the AI lifecycle and make clear that public availability of data is not consent. The floor is real.

A tool that removes identifying details before any cloud model sees them, keeps that processing on your own machine, and records what it did helps you speak to both the framework and the Act at once. That is what we built RedactLocal to do. It is built for compliance with these regimes rather than a certification of them, and the responsibility, as the Court is careful to say, stays with the lawyer.