Blog
On accuracy, AI, and Philippine law.
Notes from the team building BatasDB — on why grounded legal research beats confident guessing, and what it takes to trust an AI with the law.
Build on BatasDB
How to build your firm's internal law search engine on the BatasDB API
A developer's guide to putting grounded, effectivity-aware Philippine legal search inside your own tools — with one POST /search call. Real curl, real JSON, no scraping, no vector DB to babysit.
Legal research discipline
What Is a Case Citator, and Why Philippine Legal Research Needs One
A plain-English explanation of case citators, why US lawyers use tools like Shepard's and KeyCite, and why checking whether a case is still good law matters in Philippine legal research.
AI and the future of legal work
AI won't replace legal researchers. It'll make the good ones dangerous.
The fear that AI is coming for the lawyers misreads what legal work actually is. AI is a phenomenal librarian and a terrible advocate — with the right tech underneath, it deletes the grunt work and hands the judgment back to the person qualified to exercise it. Plus: five falsifiable questions to ask any legal AI tool.
Under the hood
The hard part isn't the AI. It's the law.
Accurate Philippine legal search isn't a model problem — it's a domain problem. The failure modes that trip up legal search, and the engineering BatasDB built to handle each: repealed-but-current provisions, short statutes losing to long cases, citations that point to half a rule.
The danger of AI legal opinions
An AI that gives legal opinions is a liability, not a lawyer.
General chatbots make case law up — and even purpose-built legal AI still hallucinates on 1 in 6 to 1 in 3 queries. Here's why, what it means for Philippine practice, and how BatasDB compares to today's Philippine legal AI tools and ChatGPT.