Posts Archive
Technical and litigation notes
Licensing Behavioral AI — Building a Revenue Model Around IP
A behavioral AI architecture creates multiple licensable IP assets: the architecture, the trained behavioral models, the ontology schema, the API, and the brand. Each can be monetized through a different licensing model. The right combination creates compounding IP moats.
IP Law for AI BuildersAPI Design as IP — Copyright, Contract, and Competition Law
After Oracle v. Google (2021 SCOTUS), API structure occupies ambiguous copyright territory. The more reliable IP protection for APIs comes from contract — developer terms — combined with trademark and competition law considerations for dominant platforms.
IP Law for AI BuildersAI Patent Prosecution Strategy for India
India's §3(k) exclusion bars patents on "computer programmes per se" — but the Ferid Allani ruling and CRI Guidelines 2016 have created a viable path for AI patents that produce a technical effect. Here is how to navigate the Indian Patent Office.
IP Law for AI BuildersPrior Art in Behavioral AI — Freedom-to-Operate Analysis
Before commercializing a behavioral AI system, you need to know whether you are infringing someone else's patent. Freedom-to-Operate (FTO) analysis for AI systems is more complex than for traditional software patents because the claim scope of behavioral AI patents is often deliberately broad.
IP Law for AI BuildersDatabase Rights and Knowledge Graph IP
Building a behavioral ontology, embedding model, or knowledge graph involves substantial investment. In the EU, that investment creates an enforceable sui generis database right lasting 15 years. In India and the US, protection relies on copyright in selection and arrangement. GDPR complicates both.
IP Law for AI BuildersTrademark Strategy for AI Product Names
A trademark protects a brand name's ability to identify the source of goods or services. For AI products with names like the WhatsApp AI agent, the legal SaaS platform, the chatbot platform, and the behavioral AI platform, the trademark analysis — registrability, class selection, and international filing strategy — varies significantly.
IP Law for AI BuildersOpen Source Licensing Strategy for AI Platforms
Choosing an open source license is one of the most consequential IP decisions an AI platform builder makes. MIT, Apache 2.0, AGPL, and the new RAIL licenses each create different obligations and protections. Getting this wrong can destroy a commercial model.
IP Law for AI BuildersCopyright in AI Code — What's Protected and What Isn't
Software copyright protects original expression — not algorithms, interfaces, or ideas. The line between what is protected and what is not is genuinely unclear for AI-assisted code, and the 2023–2025 US Copyright Office guidance has not resolved everything.
IP Law for AI BuildersDrafting Patent Claims for Multi-Engine AI Orchestration
A patent application lives or dies by its claims. For multi-engine AI orchestration systems, the challenge is writing claims specific enough to survive §101 (US) or §3(k) (India) rejections while remaining broad enough to provide meaningful competitive protection.