by Jerod Johnson | June 18, 2024

Key Takeaways from Databricks Data & AI Summit 2024

Databricks Data + AI Summit 2024 Recap

Representatives from CData's sales and marketing teams just returned from San Francisco, where we had the pleasure of representing CData Software at the Databricks Data & AI Summit. With 16,000 live and over 60,000 virtual attendees, this event had one of the largest data audiences of any event we've attended. In the expo hall, we had the opportunity to speak with hundreds of attendees and discuss how CData helps organizations with their data and AI initiatives.

In-person conversations are always enlightening, and I was particularly struck by some of the announcements and messaging from the opening keynote. You'll find my thoughts below!

Data intelligence = Democratized data + democratized AI

Early in the keynote, Ali Ghodsi, Databricks co-founder and CEO, shared several pending innovations coming to the Databricks platform. I really appreciated his definition of data intelligence: Democratized data plus democratized AI.

Democratized data means that every stakeholder in an organization has access to the data they need. Ghodsi went on to say that true data democratization means natural language access to data. He defined democratized AI as organizations and departments building AI models on their own data, not on large bespoke data sets unrelated to their tasks and initiatives.

AI for altruism

Two guests of the keynote shared how their organizations are using AI to drive progress outside of growth and profit. Brian Ames, General Motors Sr. Manager, Transformation & Enablement: Production AI and Data Products, shared General Motors' goal of zero crashes, zero emissions, zero congestion with the aim to save lives, preserve the environment, and increase comfort.

To solve for this challenge, Brian stated that the "115-year-old company that has a storied legacy in hardware must become a software company – no easy trick." General Motors is driving internal AI initiatives to build "smarter" cars to keep passengers and pedestrians safer, reduce emissions, and move towards assisted driving to keep traffic flowing during peak hours.

Fei Fei Li, Stanford University Professor and Denning Co-Director of the Stanford Institute for Human-Centered AI, shared how her team is using AI to build three-dimensional models of real-world spaces to train AI for real-world use cases. With digital clones of the real world, teams can train robot arms to perform tasks like making meals for persons with disabilities in a controlled, virtual environment without spending the time and resources to build physical prototypes.

Every company's business data is their gold mine

Jensen Huang, NVIDIA co-founder and CEO, shared the stage with Ghodsi to talk about some of their new partnering initiatives. In their session, Huang stated that, "every company's business data is their gold mine," acknowledging that organizations want and need to use their own data for their AI initiatives. One of the collaborations the duo shared was a partnership five years in the making, where Databricks will use NVIDIA GPUs to power the Photon engine, bringing accelerated data processing to all customers to help every stakeholder wrangle data faster and more efficiently, while consuming a lot less energy. With accelerated data processing, teams will be able to quickly build models and train AI to find insights to help solve problems and drive business.

Data warehouses & lakehouses are everywhere

In the hundreds of conversations we had on the expo hall floor with customers, prospects, and partners, we found that the vast majority are using a data warehouse or data lakehouse to help drive data democratization.

With business data spread across a variety of sources, clouds, and on-premises locations, organizations need the best tool to "hydrate the data lake" (a phrase commonly used by Databricks employees). CData Sync does just that, allowing you to build automated, continuous pipelines, hosted on-premises or in the cloud, for all your data.

Bonus: Live demos are the best

As a regular demo-er of CData solutions to a live audience, I was thoroughly impressed by Kasey Uhlenhuth, Databricks Sr. Product Manager, Mosaic AI Platform, and her presentation of the new Mosaic AI features. It's always valuable to see a new product or feature work in real-time (in this case, helping a fictional cookie conglomerate make better use of social media through AI-generated posts and images). Even more impressive (and something that will be difficult for AI to replace) is troubleshooting an issue with a live demo in real time before a live audience!

We'll see you in person again soon!

After the whirlwind of events in the first and second quarters, the tradeshow circuit is slowing down for the summer. No worries, though! We’ll be back this September for Workday Rising and more. Keep an eye on our blog and social media channels to see where we'll be this Fall season. It's always a pleasure to meet up with our customers and partners in person. We'd love to meet with you at our next in-person event!

In the meantime, join our CData Community to continue (or begin) your data journey!