Modern AI didn't emerge from nowhere — it was built on a series of mathematical and engineering breakthroughs that someone thought to patent. The Transformer architecture, now the backbone of every large language model from GPT to Gemini, was introduced in a 2017 paper by eight Google researchers and patented in 2019. PageRank — technically an AI-adjacent algorithm for ranking information by relevance — became the foundation of how machines understand importance on the web and seeded the field of graph-based machine learning. And CRISPR, while biological rather than computational, is the primary tool that AI-driven drug discovery companies are using to act on what models predict — the interface between AI inference and real-world genetic intervention. These are the patents researchers and engineers are building on.
Patents in this topic
More topics
Patent monitoring