How can enterprises accelerate AI adoption without sacrificing proper data sovereignty and governance? This was the central theme of a recent webinar hosted by Rakuten Symphony in collaboration with its Latin American distributor, Simply Tech.
The webinar featured Rakuten Symphony’s Padmarajan (Raj) Narayanan, global head of presales and solutions (enterprise), and Gaurav Jain, vice president of AIU and data products. Moderated by Sam Freitas, head of business development at Simply Tech, the discussion explored how Rakuten Symphony’s AI Workbench reconciles the need for rapid innovation with the strict requirements of regulated industries like telecommunications, finance, and government.
Many pioneering AI users started by running their models in the cloud. Narayanan said that as trials turned into production applications, IT realized that sensitive data was moving into these cloud-based applications and out of their physical control.
To better protect this data meant moving it in-house, said Jain. But this meant re-engineering the model and absorbing the complexity of maintaining the system, including the staff needed to run the models and the required operations software tools (AIOps, MLOps). Hyperscalers had this infrastructure and could deploy new models quickly, whereas in-house efforts could take months to develop.
Narayanan summarized the pressure that landed on the IT team: “Every enterprise leader in AI right now is living with two pressures that seem impossible to reconcile. On one side, the speed pressure: New models every quarter. Competitors shipping AI in weeks. Your board asks why you aren't already shipping. On the other side, there is a sovereignty pressure. Your data has to stay inside your borders. Regulators are watching, and every breach in the news erodes customer trust a little more.”
The discussion highlighted several additional pain points:
To succeed, Jain said, enterprise AI platforms need an integrated feature set that includes data discovery, governance, access control, model deployment, compliance auditing and lifecycle management—all working together to operate reliably.
Rakuten AI Workbench is an AI platform designed for the complete AI lifecycle across on-premises, private cloud, public cloud, hybrid and edge environments. It combines data management, governance, AI development and deployment, MLOps and operational automation on a unified platform.
Jain summarized the impact of the software: “AI Workbench is an entire toolkit with an on-prem language model, which you can use, plus it's a lifecycle with a complete toolkit that is needed for your AI journey.”
Many listeners during the webinar asked questions about AI Workbench, seeking more information about features designed to keep enterprise data under the organization’s control. Jain shared that the software can run entirely within an enterprise data center, with no connection to the internet or Rakuten Symphony servers, preventing data or models from being exposed to external organizations.
“AI Workbench is an entire toolkit with an on-prem language model, which you can use, plus it's a lifecycle with a complete toolkit that is needed for your AI journey.” – Gaurav Jain, vice president of AIU and data products, Rakuten Symphony
Key Takeaways
The enthusiastic response from the webinar audience underscores the importance of maintaining strict data security controls while also developing a flexible, scalable AI system.
Raj concluded the webinar with a reminder of what is most important in building an AI supply chain: “Eventually, models will become a commodity, but what is different is your data. How do you gather, analyze and protect your data? That becomes a major part of this exercise.”
To hear more of the discussion, watch the full webinar here: https://youtu.be/Dz_a4DQZMA0
