Open source Generative AI models generally offer varied benefits such as lower initial and maintenance costs, broader community support, faster response and speed to market, better code transparency, flexibility etc. as compared to closed source GenAI models. But at the same time, use of open source GenAI may impose varied risks such as possibility of trade secret & data leakage, output bias and hallucinations for users. Further, developers could also face copyright issues in case copyrighted data is used without permission and licensing risks for output code created etc. Multiple governments and regulatory bodies across the globe are undertaking initiatives to draft policies, standards for copyright protection, risk mitigation, data and national security and legal compliance etc. to ensure safe, secure and ethical use of Generative AI.

We will cover open & closed source Generative AI Models, Intellectual property data ownerships, risks, regulatory and legal compliance. The talk will also enlighten various key security & practical considerations in choosing open-source AI models for business workflows along with industry- led lessons learnt, best-practices and frameworks used.