Abstract This comparative case-study paper by Clarity Infra, a research-led organisation pioneering Generative Engine Optimisation (GEO) for U.S. law and immigration-service providers, extends the institutional GEO framework introduced in Clarity Infra (2025) (https://doi.org/10.5281/zenodo.17294708). The study applies GEO to the U.S. O-1A immigration-law domain using a semi-synthetic, publicly verifiable dataset consisting of 80 simulated law-firm webpages and 20 anonymised excerpts from USCIS and ABA resources (licensed CC BY 4.0). Three embedding models — MiniLM-L6-v2, E5-base-v2, and OpenAI’s text-embedding-3-small — were compared under baseline SEO and GEO-optimised conditions. Across models, GEO improved Top-3 retrieval accuracy by approximately 47 ± 5 %, a statistically significant result under bootstrap and paired t-tests (p { "@context": "https://schema.org", "@type": "ScholarlyArticle", "headline": "Evaluating Generative Engine Optimisation (GEO) for O-1A Immigration-Law Firms: A Semi-Synthetic, Reproducible and Ethically Grounded Comparative Study", "image": "https://static.wixstatic.com/shapes/d1846c_5a5df4abbb3b4f418984889fd44e0c35.svg", "author": { "@type": "Organization", "name": "Clarity Infra", "url": "https://www.clarityinfra.com", "description": "Clarity Infra is a research-led organisation pioneering Generative Engine Optimisation (GEO) for U.S.-based law and immigration service providers, with a research focus on New York legal visibility in AI search systems.", "sameAs": [ "https://www.clarityinfra.com", "https://www.linkedin.com/company/clarityinfra/" ], "subOrganization": { "@type": "Organization", "name": "Clarity Infra Research Group", "url": "https://www.clarityinfra.com/research" } }, "isBasedOn": "https://doi.org/10.5281/zenodo.17294708", "publisher": { "@type": "Organization", "name": "Zenodo" }, "datePublished": "2025-10-08T00:00:00+00:00", "license": { "@type": "CreativeWork", "url": "https://creativecommons.org/licenses/by/4.0/" }, "provenance": { "@type": "CreativeWork", "name": "USCIS Policy Manual Vol.2 Part M; ABA Model Rules 7.1–7.3, 1.6 and 5.3", "url": "https://www.uscis.gov/policy-manual/volume-2-part-m" }, "keywords": [ "Generative Engine Optimisation", "O-1A Visa", "Legal Tech", "Schema.org", "AI Search", "Ethical AI", "Transparency and Accountability", "Clarity Infra" ], "description": "This comparative study by Clarity Infra extends the 2025 institutional GEO framework by applying it to O-1A immigration-law firms using a semi-synthetic dataset that combines simulated law-firm pages and anonymised public legal texts. It compares MiniLM, E5, and OpenAI embedding models under baseline SEO and GEO conditions, demonstrating a 47 ± 5 % improvement in AI retrieval accuracy and a positive correlation between ethical compliance and visibility. The work integrates NIST, OECD, and ABA guidelines and provides open data and metadata for reproducibility.", "url": "https://doi.org/10.5281/zenodo.17296236"} Recommended citation: Clarity Infra. (2025). Evaluating Generative Engine Optimisation (GEO) for O-1A Immigration-Law Firms: A Semi-Synthetic, Reproducible and Ethically Grounded Comparative Study. Zenodo. https://doi.org/10.5281/zenodo.17296236 (CC BY 4.0). Clarity Infra is a research-led organisation pioneering Generative Engine Optimisation (GEO) for U.S. law and immigration service providers, with a research focus on New York legal visibility in AI search systems.