Feature
Like other tech conferences, there was much information shared at the recent Snowflake Data Cloud Summit 2024. Here, however, I will concentrate on two primary aspects of the summit: First, the current state of generative AI and second, the relevant use cases being implemented for GenAI. Snowflake’s tag line for the event was "AI + Data + Cloud." Clearly, as data has moved to the cloud, it is natural for AI to follow suit. This was, in fact, suggested by Jensen Huang, CEO of NVIDIA, at the event.
Keynote Presentations: How AI Transforms Snowflake
Sridhar Ramaswamy, Snowflake CEO
Sridhar Ramaswamy, now three months into his role as CEO of Snowflake, epitomizes the company’s ongoing transformation. Despite his initial background in databases, Ramaswamy brings extensive experience with AI, having worked at startups and Google. His most recent position at Snowflake involved leading Snowflake’s AI business.
Addressing the audience as "data nation," Ramaswamy highlighted Snowflake’s ambition to be the hub for data and AI, branding their cloud as the "AI Data Cloud." He shared their current scale, revealing they process five billion queries a day for enterprises, matching the volume of Google’s search business volume. In passing, he announced an open-source data catalog. This, without question, will challenge existing catalog vendors. Mirroring competitor Databricks' move toward a unified platform, Ramaswamy emphasized that Snowflake's solution avoids the pitfalls of legacy systems, which required “extensive integrations.” He stressed that GenAI too needs a catalog and centralized security and access, highlighting the enormous possibilities AI brings. This also shows that Snowflake, like Databricks, is trying to grow by taking business away from their historical data stack partners.
Ramaswamy showed strong conviction for AI and GenAI, envisioning a future where users interact with AI via natural language. He emphasized the need for data to be AI-ready, reliable and trustworthy, as Gartner advises, while also protecting privacy and managing AI costs. He highlighted the importance of the industry transitioning to enterprise-caliber AI and revealed that he has given his company a “blank check” to achieve this end, including hiring top AI talent away from Google and other leading AI companies.
As an example, he mentioned Coda Brain for collaborative workspaces, which leverages GenAI to analyze internal documents and unstructured data. This use case showcases the practical applications of GenAI in enhancing productivity and collaboration within organizations.
Jensen Huang, NVIDIA CEO
Jensen Huang emphasized the importance of making GenAI cost-effective by reducing training time. This hinges, of course, on the performance of the infrastructure and training. At the same time, Huang noted that running large models remains complex. He highlighted business service integrations, like TensorFlow and Nemo with Snowflake, stressing the necessity of moving compute to the data, given the massive data volumes involved.
When asked about opportunities for generative AI by Ramaswamy, Huang pointed to customer service as an intriguing application. Here, AI acts as the first-level support agent, fundamentally transforming subsequent stages of support. Huang explained that integrating GenAI into business processes creates a “flywheel effect,” rapidly accelerating advancements.
Huang also claimed that GenAI is advancing faster than Moore's Law, with progress doubling every six months. He asserted that the use of GenAI is fundamentally transforming businesses, encompassing both hardware and software. He urged every business to encode their processes into GenAI, highlighting NVIDIA’s own efforts to create their own internal flywheel.
Benoit Dageville, Snowflake product president
Benoit Dageville said that AI's effectiveness hinges on the quality of data, supported by elastic compute, secure models in service and collaborative AI models. He said that data platforms need to be data complete, unifying transactional and analytical data with integration and container services and efficient pipelines for both structured and unstructured data.
Snowflake is now embedding models and a large language model (LLM) into their platform service, offering to isolate them to a degree from security risks. Dageville claimed that they use fewer tokens than other LLMs. He also claimed that delivering these serverless ensures security and governance. Developers can isolate workloads, eliminate data movement and leverage applications tailored to their needs, while paying only for what they use. Without question, having an extensible data flow engine surpasses legacy systems in speed, catering to various AI models and addressing market gaps.
Christian Kleinerman, Snowflake EVP products
Christian Kleinerman talked a lot about product introductions, but he also emphasized the importance of "data on your terms," accommodating structured, semi-structured and unstructured data. An interesting new development, titled Document AI, allows users to ask questions of documents and fine-tune models, ensuring safety and governance, as was exemplified by Northern Trust. They are using it to integrate unstructured data and generate structured data from the unstructured data.
Kleinerman discussed the open-source catalog and how it is enhancing interoperability and partnering for it with ecosystems, like AWS and Microsoft. Arun Ulagaratchagan of Microsoft highlighted this point when he was brought on stage. He said the catalog fits their AI Fabric and their version of this open-source data catalog. Kleinerman also introduced their data governance offerings, allowing users to set policies, manage access and find data assets in their environment with universal search. At the same time, he discussed their marketplace that includes models and data products.
A new data classification interface offers custom classification for sensitive data, like tags within Collibra and Alation, with automated tag propagation. Users can manage access, roles and lineage, ensuring security and discoverability of data through tags. Their coming trust center provides security and compliance controls, reinforcing Snowflake's commitment to data safety.
Kleinerman suggested they are accelerating AI with Snowflake Copilot and low-code machine learning (ML) solutions. Kleinerman demonstrated how generative AI and chat features enable self-service business intelligence (BI), showcasing an inexperienced user creating a chatbot powered by Mistral Large. He emphasized here that AI is for everyone and highlighting accessibility and ease of use.
Dietmar Mauersberger, Siemens VP of data
Dietmar Mauersberger highlighted how generative AI is transforming their business and processes. Siemens has defined 450 specific use cases. By integrating the collective knowledge of the company, he said Siemens is merging the digital and physical worlds, driving efficiencies, growth and connectivity.
This approach helps them stay in contact with the products they create, extending their value to customers. This is like the premise of the book “Fusion Strategy: How Real-Time Data and AI Will Power the Industrial Future.” Siemens couldn’t achieve these ends with their legacy systems, prompting the creation of the Siemens Data Cloud. This cloud now serves 1,000 different projects, streamlining operations and enhancing their data capabilities.
See more: Effective AI Data Governance: A Strategic Ally for Success
Awinash Sinha, Zoom corporate CIO
Awinash Sinha has a keen interest in leveraging GenAI to boost sales among other use cases. While recognizing its potential, Sinha worries about the associated security concerns. His strategy emphasizes starting with desired business outcomes, particularly the phenomenal growth Zoom has experienced, and the importance of robust business operations in a subscription-based model.
Zoom's recent efforts in maturing their ML capabilities have focused on minimizing customer churn and enhancing data strategy. By breaking down data silos and defining data products that span the customer life cycle, Zoom aims to optimize critical business functions. These efforts have culminated in customer 360 insights directly in Zoom team chat, creating a multiplier effect by offering deeper insights into customer behavior across segments. They also perceive a significant opportunity identified in text-based data and mining customer data to understand lost deals. With this said, Sinha underscored that a solid foundation of privacy and data intersections is essential before fully harnessing ML and GenAI.
Sasha Jory, Hastings Direct CIO
Sasha Jory emphasized her company's commitment to being the leading digital insurer in the U.K. by leveraging data creatively to benefit customers. Jory highlighted the importance of speed and preparedness, while ensuring customer-centricity. Utilizing co-pilot technology is already enhancing customer experience and driving up the net promoter score. Jory underscored the critical role of data and the importance of better governance and controls.
By using 40-50 data points to assess risk, Hastings Direct can refine pricing strategies and identify trends, such as increases in soft tissue injuries. The company aims to enhance self-service BI with natural language processing (NLP), allowing users to ask straightforward questions, like policy counts, ultimately freeing up resources for transformational changes. Jory also values partnerships to explore innovative solutions that align with her organization’s business goals.
Cyrus Tibbs, PennyMac Financial CISO
Cyrus Tibbs is spearheading an initiative to transition their security information and event management (SIEM) system into Snowflake. By leveraging the data lake capabilities, they can store and manage more data efficiently. This move aims to empower security analysts by providing consolidated SIEM data, which is not only cost-effective, but also enhances analytical capabilities and skills.
In effect, this transition places Snowflake in direct competition with traditional SIEM vendors, like Splunk. One of the key drivers behind this shift is the ambition to use GenAI to identify patterns in log activity data. Tibbs emphasized the importance of maintaining an organized data infrastructure, aiming to use GenAI to monitor the movement of confidential data and detect unusual access patterns. This initiative underscores the need to elevate IT team data literacy to effectively analyze user behavior and reinforce data security.
Anu Jain, JPMorgan Chase CDO
Anu Jain shared that JPMorgan Chase is undergoing a significant data transformation to benefit their 82 million customers. This transformation aims to drive highly personalized customer experiences by understanding customers better and improving segmentation. Initially, AI was used to reduce risk and eliminate costs, but the focus has shifted to anticipating customer needs, leading to more effective emails and offers. Beyond personalization, JPMorgan Chase is leveraging AI for software engineering and workflow improvements. They are packaging these advancements as data sets and data products, enhancing their overall data capabilities and operational efficiency.
Hema Sundaram, Portland General Electric CIO
Hema Sundaram shared that despite only starting this year, the company already has 12 bots in production across engineering, HR and legal departments. A significant early success is in validating engineering drawings. Historically, the standard document review process took 12 days and cost $25 million annually for 5,000-6,000 designs. With generative AI, this process now takes six minutes, dramatically reducing the burdened cost, which is $1,000 per hour.
Sundaram emphasized the importance, nevertheless, of keeping humans in the loop and addressing concerns around personal information (PI) and privacy. They trust data in motion rather than storing it at rest. Additionally, they are mindful of bias in their AI implementations and are utilizing AWS and Snowflake for these projects.
Webinar
Sep
25
CMS Briefing: A Live Look at What’s Next in AI-Driven Platforms
Learn how leading organizations are using AI‑driven tools to publish faster, personalize smarter and stay secure.
Register
Webinar
Oct
15
Agentic AI Playbook Real-World Customer Service Use Cases You Can Deploy Now
Boost self-service by 30% and slash call volume by 63% with agentic AI.
Register
Webinar
On demand
Ready or Not: How Data-First Organizations Are Unlocking Agentforce Potential
Learn how to cut through the noise, activate Agentforce and build a Salesforce AI strategy that actually delivers.
Watch Now
Webinar
On demand
AI in Customer Service: Faster Resolutions, Happier Customers
Don’t let rising demand burn out your team. See how to build a smarter, more resilient support org.
Watch Now
Webinar
On demand
From Hype to High-Impact CX Strategies That Actually Scale
Turn buzzworthy AI and outsourcing trends into measurable CX wins with fresh 2025 data.
Watch Now
Webinar
On demand
Insights to Action Rethinking the Contact Center for Real Business Impact
Join our exclusive webinar to hear CX executives share their innovative strategies for transforming service delivery.
Watch Now
Gaurav Bhandan, Infosys AVP and head of data and analytics consulting
Gaurav Bhandan addressed the importance of data and AI governance in balancing the risks and returns of AI programs. Sharing Infosys research, he highlighted that 96% of companies intend to use AI technologies, including generative AI, and 56% have adopted AI in at least one business function. Goldman Sachs projects that AI investment will reach $200 billion by 2025, with a global AI market value of $1.8 trillion by 2030.
While enterprises embrace AI, Bhandan pointed out critical concerns. AI's effectiveness is only as good as the data behind it, with executives citing data usability, privacy and security as the biggest obstacles to success. AI models can become opaque black boxes, leading to trust issues. If unchecked, AI can cause significant disruptions. Therefore, ensuring trust, ethics, privacy, compliance and security in data, AI models, systems and their usage is crucial.
Bhandan suggested establishing an AI governance organization with clear policies and standards, KPIs and measures and the right tools and technology. He advocated for layers that include unified data and model metadata, smart data and modeling, data and modeling security, automated privacy controls, compliance management and usage modeling to ensure effective AI governance.
Parting Words
Snowflake Data Cloud Summit 2024 featured insights from industry leaders on the transformative power of AI and generative AI across various sectors. These leaders discussed how AI is revolutionizing business processes, enhancing customer experiences and driving efficiencies. They emphasized the importance of data quality, governance, privacy and security, with practical applications ranging from document validation to customer service and software engineering. The consensus is clear: AI's potential is vast, but its success hinges on robust governance and ethical practices.
See more: 10 Insights for Chief Data Officers From Gartner's Data & Analytics Summit
