Shadow AI is a typical approach on how most enterprises get an early taste of AI. It starts out with somebody having an idea followed by a small prototype. The public cloud is an excellent place to experiment as resources can be spun up quickly and comes with a lot of flexibility. The challenge is how do you go from prototyping to a more structured approach such as MLOps. More importantly, how do you make the move to production? Shadow AI is needed as a way to bootstrap an AI project. After all, AI is all about experimenting and comes with its failures and successes. The guests for this podcast are from run.ai; a company focused on delivering an AI platform that decouples data science workloads from the underlying hardware. Roonen Dar is the CTO and co-founder and Gijsbert Janssen van Doorn, Director of Technical Product Marketing. In the podcast we discuss why Shadow AI is needed and what you need to look out for to go all the way to production.
Roonen and Gijsbert are joined by host Stephen Foskett (Gestalt IT) and co-host Frederic Van Haren (HighFens Inc.).
A link to this podcast “Utilizing AI” episode can be found here, or click on the video below.