2026-05-05
LLM04:2025 — Data and Model Poisoning
Welcome to Securing the Stochastic : A Field Guide to the OWASP LLM Top 10, part 4 ; LLM04:2025 — Data and Model Poisoning.
Information Security · GRC Engineering · AI Security & Governance · Privacy · Open Source · Languages
I am Harsh Kahate. I work in Information Security, GRC engineering, AI security and governance, and data privacy. I build and write about practical security products, DPDPA, privacy engineering, and open-source tooling. As a passion, I teach French at Alliance Française and study languages.
Current flagship
India-first open-source PrivacyOps evidence workbench for DPDPA readiness. Code, websites, notices, consent journeys, RoPA, vendors, and fix packs.
Open SvikrutiBlog
Long-form writing on information security, GRC, AI governance, privacy, open source, and languages.
Browse the blogLanguages
A public notebook for French teaching, language study, memory, culture, and practical learning systems.
Open language notesLatest from the blog
2026-05-05
Welcome to Securing the Stochastic : A Field Guide to the OWASP LLM Top 10, part 4 ; LLM04:2025 — Data and Model Poisoning.
2026-03-22
Welcome to the part 3 of the series Securing the Stochastic — A Field Guide to the OWASP LLM Top 10.
2026-02-23
When Your AI Becomes an Accidental Whistleblower
Projects
2026-06-16
India-first open-source PrivacyOps evidence workbench for DPDPA readiness.
2026-03-10
AI-powered daily security digest that turns CVEs, security news, and AI regulation updates into persona-based Notion briefings.
2026-06-17
React and Tailwind app for creating ready-to-use FLE lesson plans for written comprehension, oral comprehension, and grammar classes.