Snowflake Inc. is dropping many big names as it leans into the hype for artificial intelligence in cloud software.
announced a partnership with Nvidia Corp.
Monday night to adapt AI models for businesses that executives say will keep proprietary data safe. Earlier in the day, Snowflake revealed a partnership with Microsoft Corp.
to bring generative AI to the company’s Data Cloud platform and link the company to OpenAI.
Nvidia, Microsoft and OpenAI are good names to drop in sales events, such as hype for generative AI revolutions, and Snowflake CEO Frank Slootman added them at the Snowflakes Summit 2023 event in Las Vegas. Nvidia CEO Jensen Huang joined Slootman in a “fireside chat”[ads1]; Monday night ahead of a keynote Tuesday morning.
“A large language model turns any computer knowledge base into an application,” Huang told the assembled audience.
“The intelligence is in the data,” Slootman added.
Snowflake’s partnership with Nvidia could provide an alternative of sorts to companies concerned about ChatGPT and proprietary information. While ChatGPT surged in popularity when it launched late last year, one complaint about the AI service is that once you input data, that data isn’t “forgotten” — it remains in the AI as part of its own learning.
Read: Snowflake expands Microsoft partnership to bring generative AI to the cloud
Engineers at Samsung Electronics Co.
found out the hard way back in April, when they pasted lines of confidential code into ChatGPT to try to get the AI to fix a coding problem – but once entered into ChatGPT, that data becomes the “property” of ChatGPT’s ” brain”. Back in May, Samsung banned the use of ChatGPT as a result.
Using Nvidia’s NeMo framework for AI developers, introduced back in September, Snowflake said its customers will be able to use the data in their Snowflake accounts to create custom large language models, or LLMs, “for advanced generative AI services, including chatbots, search and summarization,” and has “the ability to customize LLM without moving data means that proprietary information remains fully secured and managed within the Snowflake platform.”
With Snowflake’s use of NeMo, customers aren’t feeding data into a growing central AI, Manuvir Das, the head of Nvidia’s enterprise data business, told MarketWatch in an earlier press conference. He likened the process to copying and pasting: when a customer starts entering data, the customer is not using the original AI model, but a copy, which they are now modifying by adding data.
“In effect, it is now a copy of the model, and that copy belongs only to that customer,” explained Das. “Nothing that happens in that copy goes back to the original base model.”
Back in March, ServiceNow Inc.
teamed up with Nvidia on generative AI, and Das said the company is already using NeMo “to train custom models for each of their customers.” Nvidia’s Huang also showed up in person at ServiceNow’s event for the announcement.
Read: Nvidia is the only ‘beat and raise’ chip maker in the near term when it comes to AI, says analyst
Snowflake’s initial public offering was one of the largest in 2020, as an investment from Warren Buffet’s Berkshire Hathaway
and strong sales growth wooed investors and drove its valuation higher than $100 billion. Snowflake’s market capitalization was below $60 billion at Monday’s close.
For the year, Nvidia shares are up 178%, while Snowflakes is up 18.3% and Microsoft’s is up 37%. Dow Jones Industrial Average
which counts Microsoft among its 30 components, is up 1.7%, the S&P 500
has advanced 12.7% and technology-heavy Nasdaq Composite
is up 27.4 percent.