Rohit Kasbekar joined us virtually to give an overview of what generative AI is and how to use it. While AI has been used for various applications since the 1950’s, modern generative AI is trained on large data sets and leverages neural networks to generate new outputs, mimicking the patterns they have observed. He cited the main uses of AI as content creation, conversational AI (chatbots), data augmentation and creative applications. It can be applied to pharmaceuticals to speed up the development of new drugs. Some of the concerns regarding generative AI include biases in training data being reflect in AI outputs, legal and ethical concerns regarding the data used, output quality control and copyright issues. To get started using generative AI, one must assess the needs of their business to determine the best tool, fine-tune the chosen AI model with high-quality data and then refine the results with human oversight. Rohit then shared with us his 28 Day AI Challenge. This challenge is a catalogue of AI tools and each day participants choose one model to explore using a basic tutorial, allowing users to quickly gain a strong foundation in AI use.

Thank you, Rohit, for your insights into the quickly evolving world of generative AI!